From 031615ac866cc3c8f1900dd4b4aae2106ad31230 Mon Sep 17 00:00:00 2001 From: ipknHama Date: Thu, 7 Aug 2014 01:18:33 +0900 Subject: source resturcturing + CMake --- tests/CMakeLists.txt | 17 ++ tests/template/Makefile | 5 + tests/template/README.template_test | 1 + tests/template/comments.json | 1 + tests/template/comments.yml | 103 ++++++++ tests/template/delimiters.json | 1 + tests/template/delimiters.yml | 158 ++++++++++++ tests/template/interpolation.json | 1 + tests/template/interpolation.yml | 238 ++++++++++++++++++ tests/template/inverted.json | 1 + tests/template/inverted.yml | 193 ++++++++++++++ tests/template/mustachetest.cc | 33 +++ tests/template/partials.json | 1 + tests/template/partials.yml | 109 ++++++++ tests/template/sections.json | 1 + tests/template/sections.yml | 256 +++++++++++++++++++ tests/template/test.py | 29 +++ tests/template/~lambdas.json | 1 + tests/template/~lambdas.yml | 149 +++++++++++ tests/test.py | 28 +++ tests/unittest.cpp | 484 ++++++++++++++++++++++++++++++++++++ 21 files changed, 1810 insertions(+) create mode 100644 tests/CMakeLists.txt create mode 100644 tests/template/Makefile create mode 100644 tests/template/README.template_test create mode 100644 tests/template/comments.json create mode 100644 tests/template/comments.yml create mode 100644 tests/template/delimiters.json create mode 100644 tests/template/delimiters.yml create mode 100644 tests/template/interpolation.json create mode 100644 tests/template/interpolation.yml create mode 100644 tests/template/inverted.json create mode 100644 tests/template/inverted.yml create mode 100644 tests/template/mustachetest.cc create mode 100644 tests/template/partials.json create mode 100644 tests/template/partials.yml create mode 100644 tests/template/sections.json create mode 100644 tests/template/sections.yml create mode 100755 tests/template/test.py create mode 100644 tests/template/~lambdas.json create mode 100644 tests/template/~lambdas.yml create mode 100644 tests/test.py create mode 100644 tests/unittest.cpp (limited to 'tests') diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt new file mode 100644 index 0000000..45943ed --- /dev/null +++ b/tests/CMakeLists.txt @@ -0,0 +1,17 @@ +cmake_minimum_required(VERSION 2.8) +project (crow_test) + + +set(PROJECT_INCLUDE_DIR +${PROJECT_SOURCE_DIR}/include +${PROJECT_SOURCE_DIR}/http-parser +) + +set(TEST_SRCS +unittest.cpp +) + +add_executable(test ${TEST_SRCS}) +target_link_libraries(test crow) +target_link_libraries( test ${Boost_LIBRARIES} ) +set_target_properties(test PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") diff --git a/tests/template/Makefile b/tests/template/Makefile new file mode 100644 index 0000000..ad0d49b --- /dev/null +++ b/tests/template/Makefile @@ -0,0 +1,5 @@ +all: + $(CXX) -Wall -std=c++11 -g -o mustachetest mustachetest.cc +.PHONY: clean +clean: + rm -f mustachetest *.o diff --git a/tests/template/README.template_test b/tests/template/README.template_test new file mode 100644 index 0000000..6307171 --- /dev/null +++ b/tests/template/README.template_test @@ -0,0 +1 @@ +spec json/yml files from https://github.com/mustache/spec diff --git a/tests/template/comments.json b/tests/template/comments.json new file mode 100644 index 0000000..30cb927 --- /dev/null +++ b/tests/template/comments.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Comment tags represent content that should never appear in the resulting\noutput.\n\nThe tag's content may contain any substring (including newlines) EXCEPT the\nclosing delimiter.\n\nComment tags SHOULD be treated as standalone when appropriate.\n","tests":[{"name":"Inline","data":{},"expected":"1234567890","template":"12345{{! Comment Block! }}67890","desc":"Comment blocks should be removed from the template."},{"name":"Multiline","data":{},"expected":"1234567890\n","template":"12345{{!\n This is a\n multi-line comment...\n}}67890\n","desc":"Multiline comments should be permitted."},{"name":"Standalone","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n{{! Comment Block! }}\nEnd.\n","desc":"All standalone comment lines should be removed."},{"name":"Indented Standalone","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n {{! Indented Comment Block! }}\nEnd.\n","desc":"All standalone comment lines should be removed."},{"name":"Standalone Line Endings","data":{},"expected":"|\r\n|","template":"|\r\n{{! Standalone Comment }}\r\n|","desc":"\"\\r\\n\" should be considered a newline for standalone tags."},{"name":"Standalone Without Previous Line","data":{},"expected":"!","template":" {{! I'm Still Standalone }}\n!","desc":"Standalone tags should not require a newline to precede them."},{"name":"Standalone Without Newline","data":{},"expected":"!\n","template":"!\n {{! I'm Still Standalone }}","desc":"Standalone tags should not require a newline to follow them."},{"name":"Multiline Standalone","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n{{!\nSomething's going on here...\n}}\nEnd.\n","desc":"All standalone comment lines should be removed."},{"name":"Indented Multiline Standalone","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n {{!\n Something's going on here...\n }}\nEnd.\n","desc":"All standalone comment lines should be removed."},{"name":"Indented Inline","data":{},"expected":" 12 \n","template":" 12 {{! 34 }}\n","desc":"Inline comments should not strip whitespace"},{"name":"Surrounding Whitespace","data":{},"expected":"12345 67890","template":"12345 {{! Comment Block! }} 67890","desc":"Comment removal should preserve surrounding whitespace."}]} \ No newline at end of file diff --git a/tests/template/comments.yml b/tests/template/comments.yml new file mode 100644 index 0000000..7b14c7f --- /dev/null +++ b/tests/template/comments.yml @@ -0,0 +1,103 @@ +overview: | + Comment tags represent content that should never appear in the resulting + output. + + The tag's content may contain any substring (including newlines) EXCEPT the + closing delimiter. + + Comment tags SHOULD be treated as standalone when appropriate. +tests: + - name: Inline + desc: Comment blocks should be removed from the template. + data: { } + template: '12345{{! Comment Block! }}67890' + expected: '1234567890' + + - name: Multiline + desc: Multiline comments should be permitted. + data: { } + template: | + 12345{{! + This is a + multi-line comment... + }}67890 + expected: | + 1234567890 + + - name: Standalone + desc: All standalone comment lines should be removed. + data: { } + template: | + Begin. + {{! Comment Block! }} + End. + expected: | + Begin. + End. + + - name: Indented Standalone + desc: All standalone comment lines should be removed. + data: { } + template: | + Begin. + {{! Indented Comment Block! }} + End. + expected: | + Begin. + End. + + - name: Standalone Line Endings + desc: '"\r\n" should be considered a newline for standalone tags.' + data: { } + template: "|\r\n{{! Standalone Comment }}\r\n|" + expected: "|\r\n|" + + - name: Standalone Without Previous Line + desc: Standalone tags should not require a newline to precede them. + data: { } + template: " {{! I'm Still Standalone }}\n!" + expected: "!" + + - name: Standalone Without Newline + desc: Standalone tags should not require a newline to follow them. + data: { } + template: "!\n {{! I'm Still Standalone }}" + expected: "!\n" + + - name: Multiline Standalone + desc: All standalone comment lines should be removed. + data: { } + template: | + Begin. + {{! + Something's going on here... + }} + End. + expected: | + Begin. + End. + + - name: Indented Multiline Standalone + desc: All standalone comment lines should be removed. + data: { } + template: | + Begin. + {{! + Something's going on here... + }} + End. + expected: | + Begin. + End. + + - name: Indented Inline + desc: Inline comments should not strip whitespace + data: { } + template: " 12 {{! 34 }}\n" + expected: " 12 \n" + + - name: Surrounding Whitespace + desc: Comment removal should preserve surrounding whitespace. + data: { } + template: '12345 {{! Comment Block! }} 67890' + expected: '12345 67890' diff --git a/tests/template/delimiters.json b/tests/template/delimiters.json new file mode 100644 index 0000000..fcf9588 --- /dev/null +++ b/tests/template/delimiters.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Set Delimiter tags are used to change the tag delimiters for all content\nfollowing the tag in the current compilation unit.\n\nThe tag's content MUST be any two non-whitespace sequences (separated by\nwhitespace) EXCEPT an equals sign ('=') followed by the current closing\ndelimiter.\n\nSet Delimiter tags SHOULD be treated as standalone when appropriate.\n","tests":[{"name":"Pair Behavior","data":{"text":"Hey!"},"expected":"(Hey!)","template":"{{=<% %>=}}(<%text%>)","desc":"The equals sign (used on both sides) should permit delimiter changes."},{"name":"Special Characters","data":{"text":"It worked!"},"expected":"(It worked!)","template":"({{=[ ]=}}[text])","desc":"Characters with special meaning regexen should be valid delimiters."},{"name":"Sections","data":{"section":true,"data":"I got interpolated."},"expected":"[\n I got interpolated.\n |data|\n\n {{data}}\n I got interpolated.\n]\n","template":"[\n{{#section}}\n {{data}}\n |data|\n{{/section}}\n\n{{= | | =}}\n|#section|\n {{data}}\n |data|\n|/section|\n]\n","desc":"Delimiters set outside sections should persist."},{"name":"Inverted Sections","data":{"section":false,"data":"I got interpolated."},"expected":"[\n I got interpolated.\n |data|\n\n {{data}}\n I got interpolated.\n]\n","template":"[\n{{^section}}\n {{data}}\n |data|\n{{/section}}\n\n{{= | | =}}\n|^section|\n {{data}}\n |data|\n|/section|\n]\n","desc":"Delimiters set outside inverted sections should persist."},{"name":"Partial Inheritence","data":{"value":"yes"},"expected":"[ .yes. ]\n[ .yes. ]\n","template":"[ {{>include}} ]\n{{= | | =}}\n[ |>include| ]\n","desc":"Delimiters set in a parent template should not affect a partial.","partials":{"include":".{{value}}."}},{"name":"Post-Partial Behavior","data":{"value":"yes"},"expected":"[ .yes. .yes. ]\n[ .yes. .|value|. ]\n","template":"[ {{>include}} ]\n[ .{{value}}. .|value|. ]\n","desc":"Delimiters set in a partial should not affect the parent template.","partials":{"include":".{{value}}. {{= | | =}} .|value|."}},{"name":"Surrounding Whitespace","data":{},"expected":"| |","template":"| {{=@ @=}} |","desc":"Surrounding whitespace should be left untouched."},{"name":"Outlying Whitespace (Inline)","data":{},"expected":" | \n","template":" | {{=@ @=}}\n","desc":"Whitespace should be left untouched."},{"name":"Standalone Tag","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n{{=@ @=}}\nEnd.\n","desc":"Standalone lines should be removed from the template."},{"name":"Indented Standalone Tag","data":{},"expected":"Begin.\nEnd.\n","template":"Begin.\n {{=@ @=}}\nEnd.\n","desc":"Indented standalone lines should be removed from the template."},{"name":"Standalone Line Endings","data":{},"expected":"|\r\n|","template":"|\r\n{{= @ @ =}}\r\n|","desc":"\"\\r\\n\" should be considered a newline for standalone tags."},{"name":"Standalone Without Previous Line","data":{},"expected":"=","template":" {{=@ @=}}\n=","desc":"Standalone tags should not require a newline to precede them."},{"name":"Standalone Without Newline","data":{},"expected":"=\n","template":"=\n {{=@ @=}}","desc":"Standalone tags should not require a newline to follow them."},{"name":"Pair with Padding","data":{},"expected":"||","template":"|{{= @ @ =}}|","desc":"Superfluous in-tag whitespace should be ignored."}]} \ No newline at end of file diff --git a/tests/template/delimiters.yml b/tests/template/delimiters.yml new file mode 100644 index 0000000..ce80b17 --- /dev/null +++ b/tests/template/delimiters.yml @@ -0,0 +1,158 @@ +overview: | + Set Delimiter tags are used to change the tag delimiters for all content + following the tag in the current compilation unit. + + The tag's content MUST be any two non-whitespace sequences (separated by + whitespace) EXCEPT an equals sign ('=') followed by the current closing + delimiter. + + Set Delimiter tags SHOULD be treated as standalone when appropriate. +tests: + - name: Pair Behavior + desc: The equals sign (used on both sides) should permit delimiter changes. + data: { text: 'Hey!' } + template: '{{=<% %>=}}(<%text%>)' + expected: '(Hey!)' + + - name: Special Characters + desc: Characters with special meaning regexen should be valid delimiters. + data: { text: 'It worked!' } + template: '({{=[ ]=}}[text])' + expected: '(It worked!)' + + - name: Sections + desc: Delimiters set outside sections should persist. + data: { section: true, data: 'I got interpolated.' } + template: | + [ + {{#section}} + {{data}} + |data| + {{/section}} + + {{= | | =}} + |#section| + {{data}} + |data| + |/section| + ] + expected: | + [ + I got interpolated. + |data| + + {{data}} + I got interpolated. + ] + + - name: Inverted Sections + desc: Delimiters set outside inverted sections should persist. + data: { section: false, data: 'I got interpolated.' } + template: | + [ + {{^section}} + {{data}} + |data| + {{/section}} + + {{= | | =}} + |^section| + {{data}} + |data| + |/section| + ] + expected: | + [ + I got interpolated. + |data| + + {{data}} + I got interpolated. + ] + + - name: Partial Inheritence + desc: Delimiters set in a parent template should not affect a partial. + data: { value: 'yes' } + partials: + include: '.{{value}}.' + template: | + [ {{>include}} ] + {{= | | =}} + [ |>include| ] + expected: | + [ .yes. ] + [ .yes. ] + + - name: Post-Partial Behavior + desc: Delimiters set in a partial should not affect the parent template. + data: { value: 'yes' } + partials: + include: '.{{value}}. {{= | | =}} .|value|.' + template: | + [ {{>include}} ] + [ .{{value}}. .|value|. ] + expected: | + [ .yes. .yes. ] + [ .yes. .|value|. ] + + # Whitespace Sensitivity + + - name: Surrounding Whitespace + desc: Surrounding whitespace should be left untouched. + data: { } + template: '| {{=@ @=}} |' + expected: '| |' + + - name: Outlying Whitespace (Inline) + desc: Whitespace should be left untouched. + data: { } + template: " | {{=@ @=}}\n" + expected: " | \n" + + - name: Standalone Tag + desc: Standalone lines should be removed from the template. + data: { } + template: | + Begin. + {{=@ @=}} + End. + expected: | + Begin. + End. + + - name: Indented Standalone Tag + desc: Indented standalone lines should be removed from the template. + data: { } + template: | + Begin. + {{=@ @=}} + End. + expected: | + Begin. + End. + + - name: Standalone Line Endings + desc: '"\r\n" should be considered a newline for standalone tags.' + data: { } + template: "|\r\n{{= @ @ =}}\r\n|" + expected: "|\r\n|" + + - name: Standalone Without Previous Line + desc: Standalone tags should not require a newline to precede them. + data: { } + template: " {{=@ @=}}\n=" + expected: "=" + + - name: Standalone Without Newline + desc: Standalone tags should not require a newline to follow them. + data: { } + template: "=\n {{=@ @=}}" + expected: "=\n" + + # Whitespace Insensitivity + + - name: Pair with Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { } + template: '|{{= @ @ =}}|' + expected: '||' diff --git a/tests/template/interpolation.json b/tests/template/interpolation.json new file mode 100644 index 0000000..d1a1a32 --- /dev/null +++ b/tests/template/interpolation.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Interpolation tags are used to integrate dynamic content into the template.\n\nThe tag's content MUST be a non-whitespace character sequence NOT containing\nthe current closing delimiter.\n\nThis tag's content names the data to replace the tag. A single period (`.`)\nindicates that the item currently sitting atop the context stack should be\nused; otherwise, name resolution is as follows:\n 1) Split the name on periods; the first part is the name to resolve, any\n remaining parts should be retained.\n 2) Walk the context stack from top to bottom, finding the first context\n that is a) a hash containing the name as a key OR b) an object responding\n to a method with the given name.\n 3) If the context is a hash, the data is the value associated with the\n name.\n 4) If the context is an object, the data is the value returned by the\n method with the given name.\n 5) If any name parts were retained in step 1, each should be resolved\n against a context stack containing only the result from the former\n resolution. If any part fails resolution, the result should be considered\n falsey, and should interpolate as the empty string.\nData should be coerced into a string (and escaped, if appropriate) before\ninterpolation.\n\nThe Interpolation tags MUST NOT be treated as standalone.\n","tests":[{"name":"No Interpolation","data":{},"expected":"Hello from {Mustache}!\n","template":"Hello from {Mustache}!\n","desc":"Mustache-free templates should render as-is."},{"name":"Basic Interpolation","data":{"subject":"world"},"expected":"Hello, world!\n","template":"Hello, {{subject}}!\n","desc":"Unadorned tags should interpolate content into the template."},{"name":"HTML Escaping","data":{"forbidden":"& \" < >"},"expected":"These characters should be HTML escaped: & " < >\n","template":"These characters should be HTML escaped: {{forbidden}}\n","desc":"Basic interpolation should be HTML escaped."},{"name":"Triple Mustache","data":{"forbidden":"& \" < >"},"expected":"These characters should not be HTML escaped: & \" < >\n","template":"These characters should not be HTML escaped: {{{forbidden}}}\n","desc":"Triple mustaches should interpolate without HTML escaping."},{"name":"Ampersand","data":{"forbidden":"& \" < >"},"expected":"These characters should not be HTML escaped: & \" < >\n","template":"These characters should not be HTML escaped: {{&forbidden}}\n","desc":"Ampersand should interpolate without HTML escaping."},{"name":"Basic Integer Interpolation","data":{"mph":85},"expected":"\"85 miles an hour!\"","template":"\"{{mph}} miles an hour!\"","desc":"Integers should interpolate seamlessly."},{"name":"Triple Mustache Integer Interpolation","data":{"mph":85},"expected":"\"85 miles an hour!\"","template":"\"{{{mph}}} miles an hour!\"","desc":"Integers should interpolate seamlessly."},{"name":"Ampersand Integer Interpolation","data":{"mph":85},"expected":"\"85 miles an hour!\"","template":"\"{{&mph}} miles an hour!\"","desc":"Integers should interpolate seamlessly."},{"name":"Basic Decimal Interpolation","data":{"power":1.21},"expected":"\"1.21 jiggawatts!\"","template":"\"{{power}} jiggawatts!\"","desc":"Decimals should interpolate seamlessly with proper significance."},{"name":"Triple Mustache Decimal Interpolation","data":{"power":1.21},"expected":"\"1.21 jiggawatts!\"","template":"\"{{{power}}} jiggawatts!\"","desc":"Decimals should interpolate seamlessly with proper significance."},{"name":"Ampersand Decimal Interpolation","data":{"power":1.21},"expected":"\"1.21 jiggawatts!\"","template":"\"{{&power}} jiggawatts!\"","desc":"Decimals should interpolate seamlessly with proper significance."},{"name":"Basic Context Miss Interpolation","data":{},"expected":"I () be seen!","template":"I ({{cannot}}) be seen!","desc":"Failed context lookups should default to empty strings."},{"name":"Triple Mustache Context Miss Interpolation","data":{},"expected":"I () be seen!","template":"I ({{{cannot}}}) be seen!","desc":"Failed context lookups should default to empty strings."},{"name":"Ampersand Context Miss Interpolation","data":{},"expected":"I () be seen!","template":"I ({{&cannot}}) be seen!","desc":"Failed context lookups should default to empty strings."},{"name":"Dotted Names - Basic Interpolation","data":{"person":{"name":"Joe"}},"expected":"\"Joe\" == \"Joe\"","template":"\"{{person.name}}\" == \"{{#person}}{{name}}{{/person}}\"","desc":"Dotted names should be considered a form of shorthand for sections."},{"name":"Dotted Names - Triple Mustache Interpolation","data":{"person":{"name":"Joe"}},"expected":"\"Joe\" == \"Joe\"","template":"\"{{{person.name}}}\" == \"{{#person}}{{{name}}}{{/person}}\"","desc":"Dotted names should be considered a form of shorthand for sections."},{"name":"Dotted Names - Ampersand Interpolation","data":{"person":{"name":"Joe"}},"expected":"\"Joe\" == \"Joe\"","template":"\"{{&person.name}}\" == \"{{#person}}{{&name}}{{/person}}\"","desc":"Dotted names should be considered a form of shorthand for sections."},{"name":"Dotted Names - Arbitrary Depth","data":{"a":{"b":{"c":{"d":{"e":{"name":"Phil"}}}}}},"expected":"\"Phil\" == \"Phil\"","template":"\"{{a.b.c.d.e.name}}\" == \"Phil\"","desc":"Dotted names should be functional to any level of nesting."},{"name":"Dotted Names - Broken Chains","data":{"a":{}},"expected":"\"\" == \"\"","template":"\"{{a.b.c}}\" == \"\"","desc":"Any falsey value prior to the last part of the name should yield ''."},{"name":"Dotted Names - Broken Chain Resolution","data":{"a":{"b":{}},"c":{"name":"Jim"}},"expected":"\"\" == \"\"","template":"\"{{a.b.c.name}}\" == \"\"","desc":"Each part of a dotted name should resolve only against its parent."},{"name":"Dotted Names - Initial Resolution","data":{"a":{"b":{"c":{"d":{"e":{"name":"Phil"}}}}},"b":{"c":{"d":{"e":{"name":"Wrong"}}}}},"expected":"\"Phil\" == \"Phil\"","template":"\"{{#a}}{{b.c.d.e.name}}{{/a}}\" == \"Phil\"","desc":"The first part of a dotted name should resolve as any other name."},{"name":"Interpolation - Surrounding Whitespace","data":{"string":"---"},"expected":"| --- |","template":"| {{string}} |","desc":"Interpolation should not alter surrounding whitespace."},{"name":"Triple Mustache - Surrounding Whitespace","data":{"string":"---"},"expected":"| --- |","template":"| {{{string}}} |","desc":"Interpolation should not alter surrounding whitespace."},{"name":"Ampersand - Surrounding Whitespace","data":{"string":"---"},"expected":"| --- |","template":"| {{&string}} |","desc":"Interpolation should not alter surrounding whitespace."},{"name":"Interpolation - Standalone","data":{"string":"---"},"expected":" ---\n","template":" {{string}}\n","desc":"Standalone interpolation should not alter surrounding whitespace."},{"name":"Triple Mustache - Standalone","data":{"string":"---"},"expected":" ---\n","template":" {{{string}}}\n","desc":"Standalone interpolation should not alter surrounding whitespace."},{"name":"Ampersand - Standalone","data":{"string":"---"},"expected":" ---\n","template":" {{&string}}\n","desc":"Standalone interpolation should not alter surrounding whitespace."},{"name":"Interpolation With Padding","data":{"string":"---"},"expected":"|---|","template":"|{{ string }}|","desc":"Superfluous in-tag whitespace should be ignored."},{"name":"Triple Mustache With Padding","data":{"string":"---"},"expected":"|---|","template":"|{{{ string }}}|","desc":"Superfluous in-tag whitespace should be ignored."},{"name":"Ampersand With Padding","data":{"string":"---"},"expected":"|---|","template":"|{{& string }}|","desc":"Superfluous in-tag whitespace should be ignored."}]} \ No newline at end of file diff --git a/tests/template/interpolation.yml b/tests/template/interpolation.yml new file mode 100644 index 0000000..1b6fff1 --- /dev/null +++ b/tests/template/interpolation.yml @@ -0,0 +1,238 @@ +overview: | + Interpolation tags are used to integrate dynamic content into the template. + + The tag's content MUST be a non-whitespace character sequence NOT containing + the current closing delimiter. + + This tag's content names the data to replace the tag. A single period (`.`) + indicates that the item currently sitting atop the context stack should be + used; otherwise, name resolution is as follows: + 1) Split the name on periods; the first part is the name to resolve, any + remaining parts should be retained. + 2) Walk the context stack from top to bottom, finding the first context + that is a) a hash containing the name as a key OR b) an object responding + to a method with the given name. + 3) If the context is a hash, the data is the value associated with the + name. + 4) If the context is an object, the data is the value returned by the + method with the given name. + 5) If any name parts were retained in step 1, each should be resolved + against a context stack containing only the result from the former + resolution. If any part fails resolution, the result should be considered + falsey, and should interpolate as the empty string. + Data should be coerced into a string (and escaped, if appropriate) before + interpolation. + + The Interpolation tags MUST NOT be treated as standalone. +tests: + - name: No Interpolation + desc: Mustache-free templates should render as-is. + data: { } + template: | + Hello from {Mustache}! + expected: | + Hello from {Mustache}! + + - name: Basic Interpolation + desc: Unadorned tags should interpolate content into the template. + data: { subject: "world" } + template: | + Hello, {{subject}}! + expected: | + Hello, world! + + - name: HTML Escaping + desc: Basic interpolation should be HTML escaped. + data: { forbidden: '& " < >' } + template: | + These characters should be HTML escaped: {{forbidden}} + expected: | + These characters should be HTML escaped: & " < > + + - name: Triple Mustache + desc: Triple mustaches should interpolate without HTML escaping. + data: { forbidden: '& " < >' } + template: | + These characters should not be HTML escaped: {{{forbidden}}} + expected: | + These characters should not be HTML escaped: & " < > + + - name: Ampersand + desc: Ampersand should interpolate without HTML escaping. + data: { forbidden: '& " < >' } + template: | + These characters should not be HTML escaped: {{&forbidden}} + expected: | + These characters should not be HTML escaped: & " < > + + - name: Basic Integer Interpolation + desc: Integers should interpolate seamlessly. + data: { mph: 85 } + template: '"{{mph}} miles an hour!"' + expected: '"85 miles an hour!"' + + - name: Triple Mustache Integer Interpolation + desc: Integers should interpolate seamlessly. + data: { mph: 85 } + template: '"{{{mph}}} miles an hour!"' + expected: '"85 miles an hour!"' + + - name: Ampersand Integer Interpolation + desc: Integers should interpolate seamlessly. + data: { mph: 85 } + template: '"{{&mph}} miles an hour!"' + expected: '"85 miles an hour!"' + + - name: Basic Decimal Interpolation + desc: Decimals should interpolate seamlessly with proper significance. + data: { power: 1.210 } + template: '"{{power}} jiggawatts!"' + expected: '"1.21 jiggawatts!"' + + - name: Triple Mustache Decimal Interpolation + desc: Decimals should interpolate seamlessly with proper significance. + data: { power: 1.210 } + template: '"{{{power}}} jiggawatts!"' + expected: '"1.21 jiggawatts!"' + + - name: Ampersand Decimal Interpolation + desc: Decimals should interpolate seamlessly with proper significance. + data: { power: 1.210 } + template: '"{{&power}} jiggawatts!"' + expected: '"1.21 jiggawatts!"' + + # Context Misses + + - name: Basic Context Miss Interpolation + desc: Failed context lookups should default to empty strings. + data: { } + template: "I ({{cannot}}) be seen!" + expected: "I () be seen!" + + - name: Triple Mustache Context Miss Interpolation + desc: Failed context lookups should default to empty strings. + data: { } + template: "I ({{{cannot}}}) be seen!" + expected: "I () be seen!" + + - name: Ampersand Context Miss Interpolation + desc: Failed context lookups should default to empty strings. + data: { } + template: "I ({{&cannot}}) be seen!" + expected: "I () be seen!" + + # Dotted Names + + - name: Dotted Names - Basic Interpolation + desc: Dotted names should be considered a form of shorthand for sections. + data: { person: { name: 'Joe' } } + template: '"{{person.name}}" == "{{#person}}{{name}}{{/person}}"' + expected: '"Joe" == "Joe"' + + - name: Dotted Names - Triple Mustache Interpolation + desc: Dotted names should be considered a form of shorthand for sections. + data: { person: { name: 'Joe' } } + template: '"{{{person.name}}}" == "{{#person}}{{{name}}}{{/person}}"' + expected: '"Joe" == "Joe"' + + - name: Dotted Names - Ampersand Interpolation + desc: Dotted names should be considered a form of shorthand for sections. + data: { person: { name: 'Joe' } } + template: '"{{&person.name}}" == "{{#person}}{{&name}}{{/person}}"' + expected: '"Joe" == "Joe"' + + - name: Dotted Names - Arbitrary Depth + desc: Dotted names should be functional to any level of nesting. + data: + a: { b: { c: { d: { e: { name: 'Phil' } } } } } + template: '"{{a.b.c.d.e.name}}" == "Phil"' + expected: '"Phil" == "Phil"' + + - name: Dotted Names - Broken Chains + desc: Any falsey value prior to the last part of the name should yield ''. + data: + a: { } + template: '"{{a.b.c}}" == ""' + expected: '"" == ""' + + - name: Dotted Names - Broken Chain Resolution + desc: Each part of a dotted name should resolve only against its parent. + data: + a: { b: { } } + c: { name: 'Jim' } + template: '"{{a.b.c.name}}" == ""' + expected: '"" == ""' + + - name: Dotted Names - Initial Resolution + desc: The first part of a dotted name should resolve as any other name. + data: + a: { b: { c: { d: { e: { name: 'Phil' } } } } } + b: { c: { d: { e: { name: 'Wrong' } } } } + template: '"{{#a}}{{b.c.d.e.name}}{{/a}}" == "Phil"' + expected: '"Phil" == "Phil"' + + - name: Dotted Names - Context Precedence + desc: Dotted names should be resolved against former resolutions. + data: + a: { b: { } } + b: { c: 'ERROR' } + template: '{{#a}}{{b.c}}{{/a}}' + expected: '' + + # Whitespace Sensitivity + + - name: Interpolation - Surrounding Whitespace + desc: Interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: '| {{string}} |' + expected: '| --- |' + + - name: Triple Mustache - Surrounding Whitespace + desc: Interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: '| {{{string}}} |' + expected: '| --- |' + + - name: Ampersand - Surrounding Whitespace + desc: Interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: '| {{&string}} |' + expected: '| --- |' + + - name: Interpolation - Standalone + desc: Standalone interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: " {{string}}\n" + expected: " ---\n" + + - name: Triple Mustache - Standalone + desc: Standalone interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: " {{{string}}}\n" + expected: " ---\n" + + - name: Ampersand - Standalone + desc: Standalone interpolation should not alter surrounding whitespace. + data: { string: '---' } + template: " {{&string}}\n" + expected: " ---\n" + + # Whitespace Insensitivity + + - name: Interpolation With Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { string: "---" } + template: '|{{ string }}|' + expected: '|---|' + + - name: Triple Mustache With Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { string: "---" } + template: '|{{{ string }}}|' + expected: '|---|' + + - name: Ampersand With Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { string: "---" } + template: '|{{& string }}|' + expected: '|---|' diff --git a/tests/template/inverted.json b/tests/template/inverted.json new file mode 100644 index 0000000..c9b550b --- /dev/null +++ b/tests/template/inverted.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Inverted Section tags and End Section tags are used in combination to wrap a\nsection of the template.\n\nThese tags' content MUST be a non-whitespace character sequence NOT\ncontaining the current closing delimiter; each Inverted Section tag MUST be\nfollowed by an End Section tag with the same content within the same\nsection.\n\nThis tag's content names the data to replace the tag. Name resolution is as\nfollows:\n 1) Split the name on periods; the first part is the name to resolve, any\n remaining parts should be retained.\n 2) Walk the context stack from top to bottom, finding the first context\n that is a) a hash containing the name as a key OR b) an object responding\n to a method with the given name.\n 3) If the context is a hash, the data is the value associated with the\n name.\n 4) If the context is an object and the method with the given name has an\n arity of 1, the method SHOULD be called with a String containing the\n unprocessed contents of the sections; the data is the value returned.\n 5) Otherwise, the data is the value returned by calling the method with\n the given name.\n 6) If any name parts were retained in step 1, each should be resolved\n against a context stack containing only the result from the former\n resolution. If any part fails resolution, the result should be considered\n falsey, and should interpolate as the empty string.\nIf the data is not of a list type, it is coerced into a list as follows: if\nthe data is truthy (e.g. `!!data == true`), use a single-element list\ncontaining the data, otherwise use an empty list.\n\nThis section MUST NOT be rendered unless the data list is empty.\n\nInverted Section and End Section tags SHOULD be treated as standalone when\nappropriate.\n","tests":[{"name":"Falsey","data":{"boolean":false},"expected":"\"This should be rendered.\"","template":"\"{{^boolean}}This should be rendered.{{/boolean}}\"","desc":"Falsey sections should have their contents rendered."},{"name":"Truthy","data":{"boolean":true},"expected":"\"\"","template":"\"{{^boolean}}This should not be rendered.{{/boolean}}\"","desc":"Truthy sections should have their contents omitted."},{"name":"Context","data":{"context":{"name":"Joe"}},"expected":"\"\"","template":"\"{{^context}}Hi {{name}}.{{/context}}\"","desc":"Objects and hashes should behave like truthy values."},{"name":"List","data":{"list":[{"n":1},{"n":2},{"n":3}]},"expected":"\"\"","template":"\"{{^list}}{{n}}{{/list}}\"","desc":"Lists should behave like truthy values."},{"name":"Empty List","data":{"list":[]},"expected":"\"Yay lists!\"","template":"\"{{^list}}Yay lists!{{/list}}\"","desc":"Empty lists should behave like falsey values."},{"name":"Doubled","data":{"two":"second","bool":false},"expected":"* first\n* second\n* third\n","template":"{{^bool}}\n* first\n{{/bool}}\n* {{two}}\n{{^bool}}\n* third\n{{/bool}}\n","desc":"Multiple inverted sections per template should be permitted."},{"name":"Nested (Falsey)","data":{"bool":false},"expected":"| A B C D E |","template":"| A {{^bool}}B {{^bool}}C{{/bool}} D{{/bool}} E |","desc":"Nested falsey sections should have their contents rendered."},{"name":"Nested (Truthy)","data":{"bool":true},"expected":"| A E |","template":"| A {{^bool}}B {{^bool}}C{{/bool}} D{{/bool}} E |","desc":"Nested truthy sections should be omitted."},{"name":"Context Misses","data":{},"expected":"[Cannot find key 'missing'!]","template":"[{{^missing}}Cannot find key 'missing'!{{/missing}}]","desc":"Failed context lookups should be considered falsey."},{"name":"Dotted Names - Truthy","data":{"a":{"b":{"c":true}}},"expected":"\"\" == \"\"","template":"\"{{^a.b.c}}Not Here{{/a.b.c}}\" == \"\"","desc":"Dotted names should be valid for Inverted Section tags."},{"name":"Dotted Names - Falsey","data":{"a":{"b":{"c":false}}},"expected":"\"Not Here\" == \"Not Here\"","template":"\"{{^a.b.c}}Not Here{{/a.b.c}}\" == \"Not Here\"","desc":"Dotted names should be valid for Inverted Section tags."},{"name":"Dotted Names - Broken Chains","data":{"a":{}},"expected":"\"Not Here\" == \"Not Here\"","template":"\"{{^a.b.c}}Not Here{{/a.b.c}}\" == \"Not Here\"","desc":"Dotted names that cannot be resolved should be considered falsey."},{"name":"Surrounding Whitespace","data":{"boolean":false},"expected":" | \t|\t | \n","template":" | {{^boolean}}\t|\t{{/boolean}} | \n","desc":"Inverted sections should not alter surrounding whitespace."},{"name":"Internal Whitespace","data":{"boolean":false},"expected":" | \n | \n","template":" | {{^boolean}} {{! Important Whitespace }}\n {{/boolean}} | \n","desc":"Inverted should not alter internal whitespace."},{"name":"Indented Inline Sections","data":{"boolean":false},"expected":" NO\n WAY\n","template":" {{^boolean}}NO{{/boolean}}\n {{^boolean}}WAY{{/boolean}}\n","desc":"Single-line sections should not alter surrounding whitespace."},{"name":"Standalone Lines","data":{"boolean":false},"expected":"| This Is\n|\n| A Line\n","template":"| This Is\n{{^boolean}}\n|\n{{/boolean}}\n| A Line\n","desc":"Standalone lines should be removed from the template."},{"name":"Standalone Indented Lines","data":{"boolean":false},"expected":"| This Is\n|\n| A Line\n","template":"| This Is\n {{^boolean}}\n|\n {{/boolean}}\n| A Line\n","desc":"Standalone indented lines should be removed from the template."},{"name":"Standalone Line Endings","data":{"boolean":false},"expected":"|\r\n|","template":"|\r\n{{^boolean}}\r\n{{/boolean}}\r\n|","desc":"\"\\r\\n\" should be considered a newline for standalone tags."},{"name":"Standalone Without Previous Line","data":{"boolean":false},"expected":"^\n/","template":" {{^boolean}}\n^{{/boolean}}\n/","desc":"Standalone tags should not require a newline to precede them."},{"name":"Standalone Without Newline","data":{"boolean":false},"expected":"^\n/\n","template":"^{{^boolean}}\n/\n {{/boolean}}","desc":"Standalone tags should not require a newline to follow them."},{"name":"Padding","data":{"boolean":false},"expected":"|=|","template":"|{{^ boolean }}={{/ boolean }}|","desc":"Superfluous in-tag whitespace should be ignored."}]} \ No newline at end of file diff --git a/tests/template/inverted.yml b/tests/template/inverted.yml new file mode 100644 index 0000000..5f8e2b2 --- /dev/null +++ b/tests/template/inverted.yml @@ -0,0 +1,193 @@ +overview: | + Inverted Section tags and End Section tags are used in combination to wrap a + section of the template. + + These tags' content MUST be a non-whitespace character sequence NOT + containing the current closing delimiter; each Inverted Section tag MUST be + followed by an End Section tag with the same content within the same + section. + + This tag's content names the data to replace the tag. Name resolution is as + follows: + 1) Split the name on periods; the first part is the name to resolve, any + remaining parts should be retained. + 2) Walk the context stack from top to bottom, finding the first context + that is a) a hash containing the name as a key OR b) an object responding + to a method with the given name. + 3) If the context is a hash, the data is the value associated with the + name. + 4) If the context is an object and the method with the given name has an + arity of 1, the method SHOULD be called with a String containing the + unprocessed contents of the sections; the data is the value returned. + 5) Otherwise, the data is the value returned by calling the method with + the given name. + 6) If any name parts were retained in step 1, each should be resolved + against a context stack containing only the result from the former + resolution. If any part fails resolution, the result should be considered + falsey, and should interpolate as the empty string. + If the data is not of a list type, it is coerced into a list as follows: if + the data is truthy (e.g. `!!data == true`), use a single-element list + containing the data, otherwise use an empty list. + + This section MUST NOT be rendered unless the data list is empty. + + Inverted Section and End Section tags SHOULD be treated as standalone when + appropriate. +tests: + - name: Falsey + desc: Falsey sections should have their contents rendered. + data: { boolean: false } + template: '"{{^boolean}}This should be rendered.{{/boolean}}"' + expected: '"This should be rendered."' + + - name: Truthy + desc: Truthy sections should have their contents omitted. + data: { boolean: true } + template: '"{{^boolean}}This should not be rendered.{{/boolean}}"' + expected: '""' + + - name: Context + desc: Objects and hashes should behave like truthy values. + data: { context: { name: 'Joe' } } + template: '"{{^context}}Hi {{name}}.{{/context}}"' + expected: '""' + + - name: List + desc: Lists should behave like truthy values. + data: { list: [ { n: 1 }, { n: 2 }, { n: 3 } ] } + template: '"{{^list}}{{n}}{{/list}}"' + expected: '""' + + - name: Empty List + desc: Empty lists should behave like falsey values. + data: { list: [ ] } + template: '"{{^list}}Yay lists!{{/list}}"' + expected: '"Yay lists!"' + + - name: Doubled + desc: Multiple inverted sections per template should be permitted. + data: { bool: false, two: 'second' } + template: | + {{^bool}} + * first + {{/bool}} + * {{two}} + {{^bool}} + * third + {{/bool}} + expected: | + * first + * second + * third + + - name: Nested (Falsey) + desc: Nested falsey sections should have their contents rendered. + data: { bool: false } + template: "| A {{^bool}}B {{^bool}}C{{/bool}} D{{/bool}} E |" + expected: "| A B C D E |" + + - name: Nested (Truthy) + desc: Nested truthy sections should be omitted. + data: { bool: true } + template: "| A {{^bool}}B {{^bool}}C{{/bool}} D{{/bool}} E |" + expected: "| A E |" + + - name: Context Misses + desc: Failed context lookups should be considered falsey. + data: { } + template: "[{{^missing}}Cannot find key 'missing'!{{/missing}}]" + expected: "[Cannot find key 'missing'!]" + + # Dotted Names + + - name: Dotted Names - Truthy + desc: Dotted names should be valid for Inverted Section tags. + data: { a: { b: { c: true } } } + template: '"{{^a.b.c}}Not Here{{/a.b.c}}" == ""' + expected: '"" == ""' + + - name: Dotted Names - Falsey + desc: Dotted names should be valid for Inverted Section tags. + data: { a: { b: { c: false } } } + template: '"{{^a.b.c}}Not Here{{/a.b.c}}" == "Not Here"' + expected: '"Not Here" == "Not Here"' + + - name: Dotted Names - Broken Chains + desc: Dotted names that cannot be resolved should be considered falsey. + data: { a: { } } + template: '"{{^a.b.c}}Not Here{{/a.b.c}}" == "Not Here"' + expected: '"Not Here" == "Not Here"' + + # Whitespace Sensitivity + + - name: Surrounding Whitespace + desc: Inverted sections should not alter surrounding whitespace. + data: { boolean: false } + template: " | {{^boolean}}\t|\t{{/boolean}} | \n" + expected: " | \t|\t | \n" + + - name: Internal Whitespace + desc: Inverted should not alter internal whitespace. + data: { boolean: false } + template: " | {{^boolean}} {{! Important Whitespace }}\n {{/boolean}} | \n" + expected: " | \n | \n" + + - name: Indented Inline Sections + desc: Single-line sections should not alter surrounding whitespace. + data: { boolean: false } + template: " {{^boolean}}NO{{/boolean}}\n {{^boolean}}WAY{{/boolean}}\n" + expected: " NO\n WAY\n" + + - name: Standalone Lines + desc: Standalone lines should be removed from the template. + data: { boolean: false } + template: | + | This Is + {{^boolean}} + | + {{/boolean}} + | A Line + expected: | + | This Is + | + | A Line + + - name: Standalone Indented Lines + desc: Standalone indented lines should be removed from the template. + data: { boolean: false } + template: | + | This Is + {{^boolean}} + | + {{/boolean}} + | A Line + expected: | + | This Is + | + | A Line + + - name: Standalone Line Endings + desc: '"\r\n" should be considered a newline for standalone tags.' + data: { boolean: false } + template: "|\r\n{{^boolean}}\r\n{{/boolean}}\r\n|" + expected: "|\r\n|" + + - name: Standalone Without Previous Line + desc: Standalone tags should not require a newline to precede them. + data: { boolean: false } + template: " {{^boolean}}\n^{{/boolean}}\n/" + expected: "^\n/" + + - name: Standalone Without Newline + desc: Standalone tags should not require a newline to follow them. + data: { boolean: false } + template: "^{{^boolean}}\n/\n {{/boolean}}" + expected: "^\n/\n" + + # Whitespace Insensitivity + + - name: Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { boolean: false } + template: '|{{^ boolean }}={{/ boolean }}|' + expected: '|=|' diff --git a/tests/template/mustachetest.cc b/tests/template/mustachetest.cc new file mode 100644 index 0000000..c4ac2c9 --- /dev/null +++ b/tests/template/mustachetest.cc @@ -0,0 +1,33 @@ +#include +#include +#include +#include +#include "../mustache.h" +#include "../json.h" +using namespace std; +using namespace crow; +using namespace crow::mustache; + +string read_all(const string& filename) +{ + ifstream is(filename); + return {istreambuf_iterator(is), istreambuf_iterator()}; +} + +int main() +{ + auto data = json::load(read_all("data")); + auto templ = compile(read_all("template")); + auto partials = json::load(read_all("partials")); + set_loader([&](std::string name)->std::string + { + if (partials.count(name)) + { + return partials[name].s(); + } + return ""; + }); + context ctx(data); + cout << templ.render(ctx); + return 0; +} diff --git a/tests/template/partials.json b/tests/template/partials.json new file mode 100644 index 0000000..e5f21a2 --- /dev/null +++ b/tests/template/partials.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Partial tags are used to expand an external template into the current\ntemplate.\n\nThe tag's content MUST be a non-whitespace character sequence NOT containing\nthe current closing delimiter.\n\nThis tag's content names the partial to inject. Set Delimiter tags MUST NOT\naffect the parsing of a partial. The partial MUST be rendered against the\ncontext stack local to the tag. If the named partial cannot be found, the\nempty string SHOULD be used instead, as in interpolations.\n\nPartial tags SHOULD be treated as standalone when appropriate. If this tag\nis used standalone, any whitespace preceding the tag should treated as\nindentation, and prepended to each line of the partial before rendering.\n","tests":[{"name":"Basic Behavior","data":{},"expected":"\"from partial\"","template":"\"{{>text}}\"","desc":"The greater-than operator should expand to the named partial.","partials":{"text":"from partial"}},{"name":"Failed Lookup","data":{},"expected":"\"\"","template":"\"{{>text}}\"","desc":"The empty string should be used when the named partial is not found.","partials":{}},{"name":"Context","data":{"text":"content"},"expected":"\"*content*\"","template":"\"{{>partial}}\"","desc":"The greater-than operator should operate within the current context.","partials":{"partial":"*{{text}}*"}},{"name":"Recursion","data":{"content":"X","nodes":[{"content":"Y","nodes":[]}]},"expected":"X>","template":"{{>node}}","desc":"The greater-than operator should properly recurse.","partials":{"node":"{{content}}<{{#nodes}}{{>node}}{{/nodes}}>"}},{"name":"Surrounding Whitespace","data":{},"expected":"| \t|\t |","template":"| {{>partial}} |","desc":"The greater-than operator should not alter surrounding whitespace.","partials":{"partial":"\t|\t"}},{"name":"Inline Indentation","data":{"data":"|"},"expected":" | >\n>\n","template":" {{data}} {{> partial}}\n","desc":"Whitespace should be left untouched.","partials":{"partial":">\n>"}},{"name":"Standalone Line Endings","data":{},"expected":"|\r\n>|","template":"|\r\n{{>partial}}\r\n|","desc":"\"\\r\\n\" should be considered a newline for standalone tags.","partials":{"partial":">"}},{"name":"Standalone Without Previous Line","data":{},"expected":" >\n >>","template":" {{>partial}}\n>","desc":"Standalone tags should not require a newline to precede them.","partials":{"partial":">\n>"}},{"name":"Standalone Without Newline","data":{},"expected":">\n >\n >","template":">\n {{>partial}}","desc":"Standalone tags should not require a newline to follow them.","partials":{"partial":">\n>"}},{"name":"Standalone Indentation","data":{"content":"<\n->"},"expected":"\\\n |\n <\n->\n |\n/\n","template":"\\\n {{>partial}}\n/\n","desc":"Each line of the partial should be indented before rendering.","partials":{"partial":"|\n{{{content}}}\n|\n"}},{"name":"Padding Whitespace","data":{"boolean":true},"expected":"|[]|","template":"|{{> partial }}|","desc":"Superfluous in-tag whitespace should be ignored.","partials":{"partial":"[]"}}]} \ No newline at end of file diff --git a/tests/template/partials.yml b/tests/template/partials.yml new file mode 100644 index 0000000..8c41543 --- /dev/null +++ b/tests/template/partials.yml @@ -0,0 +1,109 @@ +overview: | + Partial tags are used to expand an external template into the current + template. + + The tag's content MUST be a non-whitespace character sequence NOT containing + the current closing delimiter. + + This tag's content names the partial to inject. Set Delimiter tags MUST NOT + affect the parsing of a partial. The partial MUST be rendered against the + context stack local to the tag. If the named partial cannot be found, the + empty string SHOULD be used instead, as in interpolations. + + Partial tags SHOULD be treated as standalone when appropriate. If this tag + is used standalone, any whitespace preceding the tag should treated as + indentation, and prepended to each line of the partial before rendering. +tests: + - name: Basic Behavior + desc: The greater-than operator should expand to the named partial. + data: { } + template: '"{{>text}}"' + partials: { text: 'from partial' } + expected: '"from partial"' + + - name: Failed Lookup + desc: The empty string should be used when the named partial is not found. + data: { } + template: '"{{>text}}"' + partials: { } + expected: '""' + + - name: Context + desc: The greater-than operator should operate within the current context. + data: { text: 'content' } + template: '"{{>partial}}"' + partials: { partial: '*{{text}}*' } + expected: '"*content*"' + + - name: Recursion + desc: The greater-than operator should properly recurse. + data: { content: "X", nodes: [ { content: "Y", nodes: [] } ] } + template: '{{>node}}' + partials: { node: '{{content}}<{{#nodes}}{{>node}}{{/nodes}}>' } + expected: 'X>' + + # Whitespace Sensitivity + + - name: Surrounding Whitespace + desc: The greater-than operator should not alter surrounding whitespace. + data: { } + template: '| {{>partial}} |' + partials: { partial: "\t|\t" } + expected: "| \t|\t |" + + - name: Inline Indentation + desc: Whitespace should be left untouched. + data: { data: '|' } + template: " {{data}} {{> partial}}\n" + partials: { partial: ">\n>" } + expected: " | >\n>\n" + + - name: Standalone Line Endings + desc: '"\r\n" should be considered a newline for standalone tags.' + data: { } + template: "|\r\n{{>partial}}\r\n|" + partials: { partial: ">" } + expected: "|\r\n>|" + + - name: Standalone Without Previous Line + desc: Standalone tags should not require a newline to precede them. + data: { } + template: " {{>partial}}\n>" + partials: { partial: ">\n>"} + expected: " >\n >>" + + - name: Standalone Without Newline + desc: Standalone tags should not require a newline to follow them. + data: { } + template: ">\n {{>partial}}" + partials: { partial: ">\n>" } + expected: ">\n >\n >" + + - name: Standalone Indentation + desc: Each line of the partial should be indented before rendering. + data: { content: "<\n->" } + template: | + \ + {{>partial}} + / + partials: + partial: | + | + {{{content}}} + | + expected: | + \ + | + < + -> + | + / + + # Whitespace Insensitivity + + - name: Padding Whitespace + desc: Superfluous in-tag whitespace should be ignored. + data: { boolean: true } + template: "|{{> partial }}|" + partials: { partial: "[]" } + expected: '|[]|' diff --git a/tests/template/sections.json b/tests/template/sections.json new file mode 100644 index 0000000..b0aa352 --- /dev/null +++ b/tests/template/sections.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Section tags and End Section tags are used in combination to wrap a section\nof the template for iteration\n\nThese tags' content MUST be a non-whitespace character sequence NOT\ncontaining the current closing delimiter; each Section tag MUST be followed\nby an End Section tag with the same content within the same section.\n\nThis tag's content names the data to replace the tag. Name resolution is as\nfollows:\n 1) Split the name on periods; the first part is the name to resolve, any\n remaining parts should be retained.\n 2) Walk the context stack from top to bottom, finding the first context\n that is a) a hash containing the name as a key OR b) an object responding\n to a method with the given name.\n 3) If the context is a hash, the data is the value associated with the\n name.\n 4) If the context is an object and the method with the given name has an\n arity of 1, the method SHOULD be called with a String containing the\n unprocessed contents of the sections; the data is the value returned.\n 5) Otherwise, the data is the value returned by calling the method with\n the given name.\n 6) If any name parts were retained in step 1, each should be resolved\n against a context stack containing only the result from the former\n resolution. If any part fails resolution, the result should be considered\n falsey, and should interpolate as the empty string.\nIf the data is not of a list type, it is coerced into a list as follows: if\nthe data is truthy (e.g. `!!data == true`), use a single-element list\ncontaining the data, otherwise use an empty list.\n\nFor each element in the data list, the element MUST be pushed onto the\ncontext stack, the section MUST be rendered, and the element MUST be popped\noff the context stack.\n\nSection and End Section tags SHOULD be treated as standalone when\nappropriate.\n","tests":[{"name":"Truthy","data":{"boolean":true},"expected":"\"This should be rendered.\"","template":"\"{{#boolean}}This should be rendered.{{/boolean}}\"","desc":"Truthy sections should have their contents rendered."},{"name":"Falsey","data":{"boolean":false},"expected":"\"\"","template":"\"{{#boolean}}This should not be rendered.{{/boolean}}\"","desc":"Falsey sections should have their contents omitted."},{"name":"Context","data":{"context":{"name":"Joe"}},"expected":"\"Hi Joe.\"","template":"\"{{#context}}Hi {{name}}.{{/context}}\"","desc":"Objects and hashes should be pushed onto the context stack."},{"name":"Deeply Nested Contexts","data":{"a":{"one":1},"b":{"two":2},"c":{"three":3},"d":{"four":4},"e":{"five":5}},"expected":"1\n121\n12321\n1234321\n123454321\n1234321\n12321\n121\n1\n","template":"{{#a}}\n{{one}}\n{{#b}}\n{{one}}{{two}}{{one}}\n{{#c}}\n{{one}}{{two}}{{three}}{{two}}{{one}}\n{{#d}}\n{{one}}{{two}}{{three}}{{four}}{{three}}{{two}}{{one}}\n{{#e}}\n{{one}}{{two}}{{three}}{{four}}{{five}}{{four}}{{three}}{{two}}{{one}}\n{{/e}}\n{{one}}{{two}}{{three}}{{four}}{{three}}{{two}}{{one}}\n{{/d}}\n{{one}}{{two}}{{three}}{{two}}{{one}}\n{{/c}}\n{{one}}{{two}}{{one}}\n{{/b}}\n{{one}}\n{{/a}}\n","desc":"All elements on the context stack should be accessible."},{"name":"List","data":{"list":[{"item":1},{"item":2},{"item":3}]},"expected":"\"123\"","template":"\"{{#list}}{{item}}{{/list}}\"","desc":"Lists should be iterated; list items should visit the context stack."},{"name":"Empty List","data":{"list":[]},"expected":"\"\"","template":"\"{{#list}}Yay lists!{{/list}}\"","desc":"Empty lists should behave like falsey values."},{"name":"Doubled","data":{"two":"second","bool":true},"expected":"* first\n* second\n* third\n","template":"{{#bool}}\n* first\n{{/bool}}\n* {{two}}\n{{#bool}}\n* third\n{{/bool}}\n","desc":"Multiple sections per template should be permitted."},{"name":"Nested (Truthy)","data":{"bool":true},"expected":"| A B C D E |","template":"| A {{#bool}}B {{#bool}}C{{/bool}} D{{/bool}} E |","desc":"Nested truthy sections should have their contents rendered."},{"name":"Nested (Falsey)","data":{"bool":false},"expected":"| A E |","template":"| A {{#bool}}B {{#bool}}C{{/bool}} D{{/bool}} E |","desc":"Nested falsey sections should be omitted."},{"name":"Context Misses","data":{},"expected":"[]","template":"[{{#missing}}Found key 'missing'!{{/missing}}]","desc":"Failed context lookups should be considered falsey."},{"name":"Implicit Iterator - String","data":{"list":["a","b","c","d","e"]},"expected":"\"(a)(b)(c)(d)(e)\"","template":"\"{{#list}}({{.}}){{/list}}\"","desc":"Implicit iterators should directly interpolate strings."},{"name":"Implicit Iterator - Integer","data":{"list":[1,2,3,4,5]},"expected":"\"(1)(2)(3)(4)(5)\"","template":"\"{{#list}}({{.}}){{/list}}\"","desc":"Implicit iterators should cast integers to strings and interpolate."},{"name":"Implicit Iterator - Decimal","data":{"list":[1.1,2.2,3.3,4.4,5.5]},"expected":"\"(1.1)(2.2)(3.3)(4.4)(5.5)\"","template":"\"{{#list}}({{.}}){{/list}}\"","desc":"Implicit iterators should cast decimals to strings and interpolate."},{"name":"Dotted Names - Truthy","data":{"a":{"b":{"c":true}}},"expected":"\"Here\" == \"Here\"","template":"\"{{#a.b.c}}Here{{/a.b.c}}\" == \"Here\"","desc":"Dotted names should be valid for Section tags."},{"name":"Dotted Names - Falsey","data":{"a":{"b":{"c":false}}},"expected":"\"\" == \"\"","template":"\"{{#a.b.c}}Here{{/a.b.c}}\" == \"\"","desc":"Dotted names should be valid for Section tags."},{"name":"Dotted Names - Broken Chains","data":{"a":{}},"expected":"\"\" == \"\"","template":"\"{{#a.b.c}}Here{{/a.b.c}}\" == \"\"","desc":"Dotted names that cannot be resolved should be considered falsey."},{"name":"Surrounding Whitespace","data":{"boolean":true},"expected":" | \t|\t | \n","template":" | {{#boolean}}\t|\t{{/boolean}} | \n","desc":"Sections should not alter surrounding whitespace."},{"name":"Internal Whitespace","data":{"boolean":true},"expected":" | \n | \n","template":" | {{#boolean}} {{! Important Whitespace }}\n {{/boolean}} | \n","desc":"Sections should not alter internal whitespace."},{"name":"Indented Inline Sections","data":{"boolean":true},"expected":" YES\n GOOD\n","template":" {{#boolean}}YES{{/boolean}}\n {{#boolean}}GOOD{{/boolean}}\n","desc":"Single-line sections should not alter surrounding whitespace."},{"name":"Standalone Lines","data":{"boolean":true},"expected":"| This Is\n|\n| A Line\n","template":"| This Is\n{{#boolean}}\n|\n{{/boolean}}\n| A Line\n","desc":"Standalone lines should be removed from the template."},{"name":"Indented Standalone Lines","data":{"boolean":true},"expected":"| This Is\n|\n| A Line\n","template":"| This Is\n {{#boolean}}\n|\n {{/boolean}}\n| A Line\n","desc":"Indented standalone lines should be removed from the template."},{"name":"Standalone Line Endings","data":{"boolean":true},"expected":"|\r\n|","template":"|\r\n{{#boolean}}\r\n{{/boolean}}\r\n|","desc":"\"\\r\\n\" should be considered a newline for standalone tags."},{"name":"Standalone Without Previous Line","data":{"boolean":true},"expected":"#\n/","template":" {{#boolean}}\n#{{/boolean}}\n/","desc":"Standalone tags should not require a newline to precede them."},{"name":"Standalone Without Newline","data":{"boolean":true},"expected":"#\n/\n","template":"#{{#boolean}}\n/\n {{/boolean}}","desc":"Standalone tags should not require a newline to follow them."},{"name":"Padding","data":{"boolean":true},"expected":"|=|","template":"|{{# boolean }}={{/ boolean }}|","desc":"Superfluous in-tag whitespace should be ignored."}]} \ No newline at end of file diff --git a/tests/template/sections.yml b/tests/template/sections.yml new file mode 100644 index 0000000..f62d9cb --- /dev/null +++ b/tests/template/sections.yml @@ -0,0 +1,256 @@ +overview: | + Section tags and End Section tags are used in combination to wrap a section + of the template for iteration + + These tags' content MUST be a non-whitespace character sequence NOT + containing the current closing delimiter; each Section tag MUST be followed + by an End Section tag with the same content within the same section. + + This tag's content names the data to replace the tag. Name resolution is as + follows: + 1) Split the name on periods; the first part is the name to resolve, any + remaining parts should be retained. + 2) Walk the context stack from top to bottom, finding the first context + that is a) a hash containing the name as a key OR b) an object responding + to a method with the given name. + 3) If the context is a hash, the data is the value associated with the + name. + 4) If the context is an object and the method with the given name has an + arity of 1, the method SHOULD be called with a String containing the + unprocessed contents of the sections; the data is the value returned. + 5) Otherwise, the data is the value returned by calling the method with + the given name. + 6) If any name parts were retained in step 1, each should be resolved + against a context stack containing only the result from the former + resolution. If any part fails resolution, the result should be considered + falsey, and should interpolate as the empty string. + If the data is not of a list type, it is coerced into a list as follows: if + the data is truthy (e.g. `!!data == true`), use a single-element list + containing the data, otherwise use an empty list. + + For each element in the data list, the element MUST be pushed onto the + context stack, the section MUST be rendered, and the element MUST be popped + off the context stack. + + Section and End Section tags SHOULD be treated as standalone when + appropriate. +tests: + - name: Truthy + desc: Truthy sections should have their contents rendered. + data: { boolean: true } + template: '"{{#boolean}}This should be rendered.{{/boolean}}"' + expected: '"This should be rendered."' + + - name: Falsey + desc: Falsey sections should have their contents omitted. + data: { boolean: false } + template: '"{{#boolean}}This should not be rendered.{{/boolean}}"' + expected: '""' + + - name: Context + desc: Objects and hashes should be pushed onto the context stack. + data: { context: { name: 'Joe' } } + template: '"{{#context}}Hi {{name}}.{{/context}}"' + expected: '"Hi Joe."' + + - name: Deeply Nested Contexts + desc: All elements on the context stack should be accessible. + data: + a: { one: 1 } + b: { two: 2 } + c: { three: 3 } + d: { four: 4 } + e: { five: 5 } + template: | + {{#a}} + {{one}} + {{#b}} + {{one}}{{two}}{{one}} + {{#c}} + {{one}}{{two}}{{three}}{{two}}{{one}} + {{#d}} + {{one}}{{two}}{{three}}{{four}}{{three}}{{two}}{{one}} + {{#e}} + {{one}}{{two}}{{three}}{{four}}{{five}}{{four}}{{three}}{{two}}{{one}} + {{/e}} + {{one}}{{two}}{{three}}{{four}}{{three}}{{two}}{{one}} + {{/d}} + {{one}}{{two}}{{three}}{{two}}{{one}} + {{/c}} + {{one}}{{two}}{{one}} + {{/b}} + {{one}} + {{/a}} + expected: | + 1 + 121 + 12321 + 1234321 + 123454321 + 1234321 + 12321 + 121 + 1 + + - name: List + desc: Lists should be iterated; list items should visit the context stack. + data: { list: [ { item: 1 }, { item: 2 }, { item: 3 } ] } + template: '"{{#list}}{{item}}{{/list}}"' + expected: '"123"' + + - name: Empty List + desc: Empty lists should behave like falsey values. + data: { list: [ ] } + template: '"{{#list}}Yay lists!{{/list}}"' + expected: '""' + + - name: Doubled + desc: Multiple sections per template should be permitted. + data: { bool: true, two: 'second' } + template: | + {{#bool}} + * first + {{/bool}} + * {{two}} + {{#bool}} + * third + {{/bool}} + expected: | + * first + * second + * third + + - name: Nested (Truthy) + desc: Nested truthy sections should have their contents rendered. + data: { bool: true } + template: "| A {{#bool}}B {{#bool}}C{{/bool}} D{{/bool}} E |" + expected: "| A B C D E |" + + - name: Nested (Falsey) + desc: Nested falsey sections should be omitted. + data: { bool: false } + template: "| A {{#bool}}B {{#bool}}C{{/bool}} D{{/bool}} E |" + expected: "| A E |" + + - name: Context Misses + desc: Failed context lookups should be considered falsey. + data: { } + template: "[{{#missing}}Found key 'missing'!{{/missing}}]" + expected: "[]" + + # Implicit Iterators + + - name: Implicit Iterator - String + desc: Implicit iterators should directly interpolate strings. + data: + list: [ 'a', 'b', 'c', 'd', 'e' ] + template: '"{{#list}}({{.}}){{/list}}"' + expected: '"(a)(b)(c)(d)(e)"' + + - name: Implicit Iterator - Integer + desc: Implicit iterators should cast integers to strings and interpolate. + data: + list: [ 1, 2, 3, 4, 5 ] + template: '"{{#list}}({{.}}){{/list}}"' + expected: '"(1)(2)(3)(4)(5)"' + + - name: Implicit Iterator - Decimal + desc: Implicit iterators should cast decimals to strings and interpolate. + data: + list: [ 1.10, 2.20, 3.30, 4.40, 5.50 ] + template: '"{{#list}}({{.}}){{/list}}"' + expected: '"(1.1)(2.2)(3.3)(4.4)(5.5)"' + + # Dotted Names + + - name: Dotted Names - Truthy + desc: Dotted names should be valid for Section tags. + data: { a: { b: { c: true } } } + template: '"{{#a.b.c}}Here{{/a.b.c}}" == "Here"' + expected: '"Here" == "Here"' + + - name: Dotted Names - Falsey + desc: Dotted names should be valid for Section tags. + data: { a: { b: { c: false } } } + template: '"{{#a.b.c}}Here{{/a.b.c}}" == ""' + expected: '"" == ""' + + - name: Dotted Names - Broken Chains + desc: Dotted names that cannot be resolved should be considered falsey. + data: { a: { } } + template: '"{{#a.b.c}}Here{{/a.b.c}}" == ""' + expected: '"" == ""' + + # Whitespace Sensitivity + + - name: Surrounding Whitespace + desc: Sections should not alter surrounding whitespace. + data: { boolean: true } + template: " | {{#boolean}}\t|\t{{/boolean}} | \n" + expected: " | \t|\t | \n" + + - name: Internal Whitespace + desc: Sections should not alter internal whitespace. + data: { boolean: true } + template: " | {{#boolean}} {{! Important Whitespace }}\n {{/boolean}} | \n" + expected: " | \n | \n" + + - name: Indented Inline Sections + desc: Single-line sections should not alter surrounding whitespace. + data: { boolean: true } + template: " {{#boolean}}YES{{/boolean}}\n {{#boolean}}GOOD{{/boolean}}\n" + expected: " YES\n GOOD\n" + + - name: Standalone Lines + desc: Standalone lines should be removed from the template. + data: { boolean: true } + template: | + | This Is + {{#boolean}} + | + {{/boolean}} + | A Line + expected: | + | This Is + | + | A Line + + - name: Indented Standalone Lines + desc: Indented standalone lines should be removed from the template. + data: { boolean: true } + template: | + | This Is + {{#boolean}} + | + {{/boolean}} + | A Line + expected: | + | This Is + | + | A Line + + - name: Standalone Line Endings + desc: '"\r\n" should be considered a newline for standalone tags.' + data: { boolean: true } + template: "|\r\n{{#boolean}}\r\n{{/boolean}}\r\n|" + expected: "|\r\n|" + + - name: Standalone Without Previous Line + desc: Standalone tags should not require a newline to precede them. + data: { boolean: true } + template: " {{#boolean}}\n#{{/boolean}}\n/" + expected: "#\n/" + + - name: Standalone Without Newline + desc: Standalone tags should not require a newline to follow them. + data: { boolean: true } + template: "#{{#boolean}}\n/\n {{/boolean}}" + expected: "#\n/\n" + + # Whitespace Insensitivity + + - name: Padding + desc: Superfluous in-tag whitespace should be ignored. + data: { boolean: true } + template: '|{{# boolean }}={{/ boolean }}|' + expected: '|=|' diff --git a/tests/template/test.py b/tests/template/test.py new file mode 100755 index 0000000..22fcca0 --- /dev/null +++ b/tests/template/test.py @@ -0,0 +1,29 @@ +#!/usr/bin/env python +import glob +import json +import os +import subprocess +for testfile in glob.glob("*.json"): + testdoc = json.load(open(testfile)) + for test in testdoc["tests"]: + if "lambda" in test["data"]: + continue + open('data', 'w').write(json.dumps(test["data"])) + open('template', 'w').write(test["template"]) + if "partials" in test: + open('partials', 'w').write(json.dumps(test["partials"])) + else: + open('partials', 'w').write("{}") + ret = subprocess.check_output("./mustachetest") + print testfile, test["name"] + if ret != test["expected"]: + if 'partials' in test: + print 'partials:', json.dumps(test["partials"]) + print json.dumps(test["data"]) + print test["template"] + print 'Expected:',repr(test["expected"]) + print 'Actual:',repr(ret) + assert ret == test["expected"] + os.unlink('data') + os.unlink('template') + os.unlink('partials') diff --git a/tests/template/~lambdas.json b/tests/template/~lambdas.json new file mode 100644 index 0000000..3c58bf8 --- /dev/null +++ b/tests/template/~lambdas.json @@ -0,0 +1 @@ +{"__ATTN__":"Do not edit this file; changes belong in the appropriate YAML file.","overview":"Lambdas are a special-cased data type for use in interpolations and\nsections.\n\nWhen used as the data value for an Interpolation tag, the lambda MUST be\ntreatable as an arity 0 function, and invoked as such. The returned value\nMUST be rendered against the default delimiters, then interpolated in place\nof the lambda.\n\nWhen used as the data value for a Section tag, the lambda MUST be treatable\nas an arity 1 function, and invoked as such (passing a String containing the\nunprocessed section contents). The returned value MUST be rendered against\nthe current delimiters, then interpolated in place of the section.\n","tests":[{"name":"Interpolation","data":{"lambda":{"php":"return \"world\";","clojure":"(fn [] \"world\")","__tag__":"code","perl":"sub { \"world\" }","python":"lambda: \"world\"","ruby":"proc { \"world\" }","js":"function() { return \"world\" }"}},"expected":"Hello, world!","template":"Hello, {{lambda}}!","desc":"A lambda's return value should be interpolated."},{"name":"Interpolation - Expansion","data":{"planet":"world","lambda":{"php":"return \"{{planet}}\";","clojure":"(fn [] \"{{planet}}\")","__tag__":"code","perl":"sub { \"{{planet}}\" }","python":"lambda: \"{{planet}}\"","ruby":"proc { \"{{planet}}\" }","js":"function() { return \"{{planet}}\" }"}},"expected":"Hello, world!","template":"Hello, {{lambda}}!","desc":"A lambda's return value should be parsed."},{"name":"Interpolation - Alternate Delimiters","data":{"planet":"world","lambda":{"php":"return \"|planet| => {{planet}}\";","clojure":"(fn [] \"|planet| => {{planet}}\")","__tag__":"code","perl":"sub { \"|planet| => {{planet}}\" }","python":"lambda: \"|planet| => {{planet}}\"","ruby":"proc { \"|planet| => {{planet}}\" }","js":"function() { return \"|planet| => {{planet}}\" }"}},"expected":"Hello, (|planet| => world)!","template":"{{= | | =}}\nHello, (|&lambda|)!","desc":"A lambda's return value should parse with the default delimiters."},{"name":"Interpolation - Multiple Calls","data":{"lambda":{"php":"global $calls; return ++$calls;","clojure":"(def g (atom 0)) (fn [] (swap! g inc))","__tag__":"code","perl":"sub { no strict; $calls += 1 }","python":"lambda: globals().update(calls=globals().get(\"calls\",0)+1) or calls","ruby":"proc { $calls ||= 0; $calls += 1 }","js":"function() { return (g=(function(){return this})()).calls=(g.calls||0)+1 }"}},"expected":"1 == 2 == 3","template":"{{lambda}} == {{{lambda}}} == {{lambda}}","desc":"Interpolated lambdas should not be cached."},{"name":"Escaping","data":{"lambda":{"php":"return \">\";","clojure":"(fn [] \">\")","__tag__":"code","perl":"sub { \">\" }","python":"lambda: \">\"","ruby":"proc { \">\" }","js":"function() { return \">\" }"}},"expected":"<>>","template":"<{{lambda}}{{{lambda}}}","desc":"Lambda results should be appropriately escaped."},{"name":"Section","data":{"x":"Error!","lambda":{"php":"return ($text == \"{{x}}\") ? \"yes\" : \"no\";","clojure":"(fn [text] (if (= text \"{{x}}\") \"yes\" \"no\"))","__tag__":"code","perl":"sub { $_[0] eq \"{{x}}\" ? \"yes\" : \"no\" }","python":"lambda text: text == \"{{x}}\" and \"yes\" or \"no\"","ruby":"proc { |text| text == \"{{x}}\" ? \"yes\" : \"no\" }","js":"function(txt) { return (txt == \"{{x}}\" ? \"yes\" : \"no\") }"}},"expected":"","template":"<{{#lambda}}{{x}}{{/lambda}}>","desc":"Lambdas used for sections should receive the raw section string."},{"name":"Section - Expansion","data":{"planet":"Earth","lambda":{"php":"return $text . \"{{planet}}\" . $text;","clojure":"(fn [text] (str text \"{{planet}}\" text))","__tag__":"code","perl":"sub { $_[0] . \"{{planet}}\" . $_[0] }","python":"lambda text: \"%s{{planet}}%s\" % (text, text)","ruby":"proc { |text| \"#{text}{{planet}}#{text}\" }","js":"function(txt) { return txt + \"{{planet}}\" + txt }"}},"expected":"<-Earth->","template":"<{{#lambda}}-{{/lambda}}>","desc":"Lambdas used for sections should have their results parsed."},{"name":"Section - Alternate Delimiters","data":{"planet":"Earth","lambda":{"php":"return $text . \"{{planet}} => |planet|\" . $text;","clojure":"(fn [text] (str text \"{{planet}} => |planet|\" text))","__tag__":"code","perl":"sub { $_[0] . \"{{planet}} => |planet|\" . $_[0] }","python":"lambda text: \"%s{{planet}} => |planet|%s\" % (text, text)","ruby":"proc { |text| \"#{text}{{planet}} => |planet|#{text}\" }","js":"function(txt) { return txt + \"{{planet}} => |planet|\" + txt }"}},"expected":"<-{{planet}} => Earth->","template":"{{= | | =}}<|#lambda|-|/lambda|>","desc":"Lambdas used for sections should parse with the current delimiters."},{"name":"Section - Multiple Calls","data":{"lambda":{"php":"return \"__\" . $text . \"__\";","clojure":"(fn [text] (str \"__\" text \"__\"))","__tag__":"code","perl":"sub { \"__\" . $_[0] . \"__\" }","python":"lambda text: \"__%s__\" % (text)","ruby":"proc { |text| \"__#{text}__\" }","js":"function(txt) { return \"__\" + txt + \"__\" }"}},"expected":"__FILE__ != __LINE__","template":"{{#lambda}}FILE{{/lambda}} != {{#lambda}}LINE{{/lambda}}","desc":"Lambdas used for sections should not be cached."},{"name":"Inverted Section","data":{"static":"static","lambda":{"php":"return false;","clojure":"(fn [text] false)","__tag__":"code","perl":"sub { 0 }","python":"lambda text: 0","ruby":"proc { |text| false }","js":"function(txt) { return false }"}},"expected":"<>","template":"<{{^lambda}}{{static}}{{/lambda}}>","desc":"Lambdas used for inverted sections should be considered truthy."}]} \ No newline at end of file diff --git a/tests/template/~lambdas.yml b/tests/template/~lambdas.yml new file mode 100644 index 0000000..b9fb4d0 --- /dev/null +++ b/tests/template/~lambdas.yml @@ -0,0 +1,149 @@ +overview: | + Lambdas are a special-cased data type for use in interpolations and + sections. + + When used as the data value for an Interpolation tag, the lambda MUST be + treatable as an arity 0 function, and invoked as such. The returned value + MUST be rendered against the default delimiters, then interpolated in place + of the lambda. + + When used as the data value for a Section tag, the lambda MUST be treatable + as an arity 1 function, and invoked as such (passing a String containing the + unprocessed section contents). The returned value MUST be rendered against + the current delimiters, then interpolated in place of the section. +tests: + - name: Interpolation + desc: A lambda's return value should be interpolated. + data: + lambda: !code + ruby: 'proc { "world" }' + perl: 'sub { "world" }' + js: 'function() { return "world" }' + php: 'return "world";' + python: 'lambda: "world"' + clojure: '(fn [] "world")' + template: "Hello, {{lambda}}!" + expected: "Hello, world!" + + - name: Interpolation - Expansion + desc: A lambda's return value should be parsed. + data: + planet: "world" + lambda: !code + ruby: 'proc { "{{planet}}" }' + perl: 'sub { "{{planet}}" }' + js: 'function() { return "{{planet}}" }' + php: 'return "{{planet}}";' + python: 'lambda: "{{planet}}"' + clojure: '(fn [] "{{planet}}")' + template: "Hello, {{lambda}}!" + expected: "Hello, world!" + + - name: Interpolation - Alternate Delimiters + desc: A lambda's return value should parse with the default delimiters. + data: + planet: "world" + lambda: !code + ruby: 'proc { "|planet| => {{planet}}" }' + perl: 'sub { "|planet| => {{planet}}" }' + js: 'function() { return "|planet| => {{planet}}" }' + php: 'return "|planet| => {{planet}}";' + python: 'lambda: "|planet| => {{planet}}"' + clojure: '(fn [] "|planet| => {{planet}}")' + template: "{{= | | =}}\nHello, (|&lambda|)!" + expected: "Hello, (|planet| => world)!" + + - name: Interpolation - Multiple Calls + desc: Interpolated lambdas should not be cached. + data: + lambda: !code + ruby: 'proc { $calls ||= 0; $calls += 1 }' + perl: 'sub { no strict; $calls += 1 }' + js: 'function() { return (g=(function(){return this})()).calls=(g.calls||0)+1 }' + php: 'global $calls; return ++$calls;' + python: 'lambda: globals().update(calls=globals().get("calls",0)+1) or calls' + clojure: '(def g (atom 0)) (fn [] (swap! g inc))' + template: '{{lambda}} == {{{lambda}}} == {{lambda}}' + expected: '1 == 2 == 3' + + - name: Escaping + desc: Lambda results should be appropriately escaped. + data: + lambda: !code + ruby: 'proc { ">" }' + perl: 'sub { ">" }' + js: 'function() { return ">" }' + php: 'return ">";' + python: 'lambda: ">"' + clojure: '(fn [] ">")' + template: "<{{lambda}}{{{lambda}}}" + expected: "<>>" + + - name: Section + desc: Lambdas used for sections should receive the raw section string. + data: + x: 'Error!' + lambda: !code + ruby: 'proc { |text| text == "{{x}}" ? "yes" : "no" }' + perl: 'sub { $_[0] eq "{{x}}" ? "yes" : "no" }' + js: 'function(txt) { return (txt == "{{x}}" ? "yes" : "no") }' + php: 'return ($text == "{{x}}") ? "yes" : "no";' + python: 'lambda text: text == "{{x}}" and "yes" or "no"' + clojure: '(fn [text] (if (= text "{{x}}") "yes" "no"))' + template: "<{{#lambda}}{{x}}{{/lambda}}>" + expected: "" + + - name: Section - Expansion + desc: Lambdas used for sections should have their results parsed. + data: + planet: "Earth" + lambda: !code + ruby: 'proc { |text| "#{text}{{planet}}#{text}" }' + perl: 'sub { $_[0] . "{{planet}}" . $_[0] }' + js: 'function(txt) { return txt + "{{planet}}" + txt }' + php: 'return $text . "{{planet}}" . $text;' + python: 'lambda text: "%s{{planet}}%s" % (text, text)' + clojure: '(fn [text] (str text "{{planet}}" text))' + template: "<{{#lambda}}-{{/lambda}}>" + expected: "<-Earth->" + + - name: Section - Alternate Delimiters + desc: Lambdas used for sections should parse with the current delimiters. + data: + planet: "Earth" + lambda: !code + ruby: 'proc { |text| "#{text}{{planet}} => |planet|#{text}" }' + perl: 'sub { $_[0] . "{{planet}} => |planet|" . $_[0] }' + js: 'function(txt) { return txt + "{{planet}} => |planet|" + txt }' + php: 'return $text . "{{planet}} => |planet|" . $text;' + python: 'lambda text: "%s{{planet}} => |planet|%s" % (text, text)' + clojure: '(fn [text] (str text "{{planet}} => |planet|" text))' + template: "{{= | | =}}<|#lambda|-|/lambda|>" + expected: "<-{{planet}} => Earth->" + + - name: Section - Multiple Calls + desc: Lambdas used for sections should not be cached. + data: + lambda: !code + ruby: 'proc { |text| "__#{text}__" }' + perl: 'sub { "__" . $_[0] . "__" }' + js: 'function(txt) { return "__" + txt + "__" }' + php: 'return "__" . $text . "__";' + python: 'lambda text: "__%s__" % (text)' + clojure: '(fn [text] (str "__" text "__"))' + template: '{{#lambda}}FILE{{/lambda}} != {{#lambda}}LINE{{/lambda}}' + expected: '__FILE__ != __LINE__' + + - name: Inverted Section + desc: Lambdas used for inverted sections should be considered truthy. + data: + static: 'static' + lambda: !code + ruby: 'proc { |text| false }' + perl: 'sub { 0 }' + js: 'function(txt) { return false }' + php: 'return false;' + python: 'lambda text: 0' + clojure: '(fn [text] false)' + template: "<{{^lambda}}{{static}}{{/lambda}}>" + expected: "<>" diff --git a/tests/test.py b/tests/test.py new file mode 100644 index 0000000..e2bf651 --- /dev/null +++ b/tests/test.py @@ -0,0 +1,28 @@ +import urllib +assert "Hello World!" == urllib.urlopen('http://localhost:18080').read() +assert "About Crow example." == urllib.urlopen('http://localhost:18080/about').read() +assert 404 == urllib.urlopen('http://localhost:18080/list').getcode() +assert "3 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/3').read() +assert "100 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/100').read() +assert 400 == urllib.urlopen('http://localhost:18080/hello/500').getcode() +assert "3" == urllib.urlopen('http://localhost:18080/add_json', data='{"a":1,"b":2}').read() +assert "3" == urllib.urlopen('http://localhost:18080/add/1/2').read() + +# test persistent connection +import socket +import time +s = socket.socket() +s.connect(('localhost', 18080)) +for i in xrange(10): + s.send('''GET / HTTP/1.1 +Host: localhost\r\n\r\n'''); + assert 'Hello World!' in s.recv(1024) + +# test timeout +s = socket.socket() +s.connect(('localhost', 18080)) +print 'ERROR REQUEST' +s.send('''GET / HTTP/1.1 +hHhHHefhwjkefhklwejfklwejf +''') +print s.recv(1024) diff --git a/tests/unittest.cpp b/tests/unittest.cpp new file mode 100644 index 0000000..918d027 --- /dev/null +++ b/tests/unittest.cpp @@ -0,0 +1,484 @@ +//#define CROW_ENABLE_LOGGING +#define CROW_ENABLE_DEBUG +#include +#include +#include "routing.h" +#include "utility.h" +#include "crow.h" +#include "json.h" +#include "mustache.h" +using namespace std; +using namespace crow; + +struct Test { Test(); virtual void test() = 0; }; +vector tests; +Test::Test() { tests.push_back(this); } + +bool failed__ = false; +void error_print() +{ + cerr << endl; +} + +template +void error_print(const A& a, Args...args) +{ + cerr< +void fail(Args...args) { error_print(args...);failed__ = true; } + +#define ASSERT_TRUE(x) if (!(x)) fail(__FILE__ ":", __LINE__, ": Assert fail: expected ", #x, " is true, at " __FILE__ ":",__LINE__) +#define ASSERT_EQUAL(a, b) if ((a) != (b)) fail(__FILE__ ":", __LINE__, ": Assert fail: expected ", (a), " actual " , (b), ", " #a " == " #b ", at " __FILE__ ":",__LINE__) +#define ASSERT_NOTEQUAL(a, b) if ((a) == (b)) fail(__FILE__ ":", __LINE__, ": Assert fail: not expected ", (a), ", " #a " != " #b ", at " __FILE__ ":",__LINE__) +#define ASSERT_THROW(x) \ + try \ + { \ + x; \ + fail(__FILE__ ":", __LINE__, ": Assert fail: exception should be thrown"); \ + } \ + catch(std::exception&) \ + { \ + } + + + +#define TEST(x) struct test##x:public Test{void test();}x##_; \ + void test##x::test() +#define DISABLE_TEST(x) struct test##x{void test();}x##_; \ + void test##x::test() + +TEST(Rule) +{ + TaggedRule<> r("/http/"); + r.name("abc"); + + // empty handler - fail to validate + try + { + r.validate(); + fail("empty handler should fail to validate"); + } + catch(runtime_error& e) + { + } + + int x = 0; + + // registering handler + r([&x]{x = 1;return "";}); + + r.validate(); + + response res; + + // executing handler + ASSERT_EQUAL(0, x); + r.handle(request(), res, routing_params()); + ASSERT_EQUAL(1, x); + + // registering handler with request argument + r([&x](const crow::request&){x = 2;return "";}); + + r.validate(); + + // executing handler + ASSERT_EQUAL(1, x); + r.handle(request(), res, routing_params()); + ASSERT_EQUAL(2, x); +} + +TEST(ParameterTagging) +{ + static_assert(black_magic::is_valid(""), "valid url"); + static_assert(!black_magic::is_valid(""), "invalid url"); + static_assert(!black_magic::is_valid("nt>"), "invalid url"); + ASSERT_EQUAL(1, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(2, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(3, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(3, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(4, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(4, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(5, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(6*6+6+1, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(6*6+6+2, black_magic::get_parameter_tag("")); + ASSERT_EQUAL(6*6+6*3+2, black_magic::get_parameter_tag("")); + + // url definition parsed in compile time, build into *one number*, and given to template argument + static_assert(std::is_same, black_magic::arguments<6*6+6*3+2>::type>::value, "tag to type container"); +} + +TEST(RoutingTest) +{ + Crow app; + int A{}; + uint32_t B{}; + double C{}; + string D{}; + string E{}; + + CROW_ROUTE(app, "/0/") + ([&](uint32_t b){ + B = b; + return "OK"; + }); + + CROW_ROUTE(app, "/1//") + ([&](int a, uint32_t b){ + A = a; B = b; + return "OK"; + }); + + CROW_ROUTE(app, "/4////") + ([&](int a, uint32_t b, double c, string d){ + A = a; B = b; C = c; D = d; + return "OK"; + }); + + CROW_ROUTE(app, "/5/////") + ([&](int a, uint32_t b, double c, string d, string e){ + A = a; B = b; C = c; D = d; E = e; + return "OK"; + }); + + app.validate(); + //app.debug_print(); + { + request req; + response res; + + req.url = "/-1"; + + app.handle(req, res); + + ASSERT_EQUAL(404, res.code); + } + + { + request req; + response res; + + req.url = "/0/1001999"; + + app.handle(req, res); + + ASSERT_EQUAL(200, res.code); + + ASSERT_EQUAL(1001999, B); + } + + { + request req; + response res; + + req.url = "/1/-100/1999"; + + app.handle(req, res); + + ASSERT_EQUAL(200, res.code); + + ASSERT_EQUAL(-100, A); + ASSERT_EQUAL(1999, B); + } + { + request req; + response res; + + req.url = "/4/5000/3/-2.71828/hellhere"; + req.headers["TestHeader"] = "Value"; + + app.handle(req, res); + + ASSERT_EQUAL(200, res.code); + + ASSERT_EQUAL(5000, A); + ASSERT_EQUAL(3, B); + ASSERT_EQUAL(-2.71828, C); + ASSERT_EQUAL("hellhere", D); + } + { + request req; + response res; + + req.url = "/5/-5/999/3.141592/hello_there/a/b/c/d"; + req.headers["TestHeader"] = "Value"; + + app.handle(req, res); + + ASSERT_EQUAL(200, res.code); + + ASSERT_EQUAL(-5, A); + ASSERT_EQUAL(999, B); + ASSERT_EQUAL(3.141592, C); + ASSERT_EQUAL("hello_there", D); + ASSERT_EQUAL("a/b/c/d", E); + } +} + +TEST(simple_response_routing_params) +{ + ASSERT_EQUAL(100, response(100).code); + ASSERT_EQUAL(200, response("Hello there").code); + ASSERT_EQUAL(500, response(500, "Internal Error?").code); + + routing_params rp; + rp.int_params.push_back(1); + rp.int_params.push_back(5); + rp.uint_params.push_back(2); + rp.double_params.push_back(3); + rp.string_params.push_back("hello"); + ASSERT_EQUAL(1, rp.get(0)); + ASSERT_EQUAL(5, rp.get(1)); + ASSERT_EQUAL(2, rp.get(0)); + ASSERT_EQUAL(3, rp.get(0)); + ASSERT_EQUAL("hello", rp.get(0)); +} + +TEST(handler_with_response) +{ + Crow app; + CROW_ROUTE(app, "/")([](const crow::request&, crow::response&) + { + }); +} + +TEST(server_handling_error_request) +{ + static char buf[2048]; + Crow app; + CROW_ROUTE(app, "/")([]{return "A";}); + Server server(&app, 45451); + auto _ = async(launch::async, [&]{server.run();}); + std::string sendmsg = "POX"; + asio::io_service is; + { + asio::ip::tcp::socket c(is); + c.connect(asio::ip::tcp::endpoint(asio::ip::address::from_string("127.0.0.1"), 45451)); + + + c.send(asio::buffer(sendmsg)); + + try + { + c.receive(asio::buffer(buf, 2048)); + fail(); + } + catch(std::exception& e) + { + //std::cerr << e.what() << std::endl; + } + } + server.stop(); +} + +TEST(multi_server) +{ + static char buf[2048]; + Crow app1, app2; + CROW_ROUTE(app1, "/")([]{return "A";}); + CROW_ROUTE(app2, "/")([]{return "B";}); + + Server server1(&app1, 45451); + Server server2(&app2, 45452); + + auto _ = async(launch::async, [&]{server1.run();}); + auto _2 = async(launch::async, [&]{server2.run();}); + + std::string sendmsg = "POST /\r\nContent-Length:3\r\nX-HeaderTest: 123\r\n\r\nA=B\r\n"; + asio::io_service is; + { + asio::ip::tcp::socket c(is); + c.connect(asio::ip::tcp::endpoint(asio::ip::address::from_string("127.0.0.1"), 45451)); + + + c.send(asio::buffer(sendmsg)); + + size_t recved = c.receive(asio::buffer(buf, 2048)); + ASSERT_EQUAL('A', buf[recved-1]); + } + + { + asio::ip::tcp::socket c(is); + c.connect(asio::ip::tcp::endpoint(asio::ip::address::from_string("127.0.0.1"), 45452)); + + for(auto ch:sendmsg) + { + char buf[1] = {ch}; + c.send(asio::buffer(buf)); + } + + size_t recved = c.receive(asio::buffer(buf, 2048)); + ASSERT_EQUAL('B', buf[recved-1]); + } + + server1.stop(); + server2.stop(); +} + +TEST(json_read) +{ + { + const char* json_error_tests[] = + { + "{} 3", "{{}", "{3}", + "3.4.5", "+3", "3-2", "00", "03", "1e3e3", "1e+.3", + "nll", "f", "t", + "{\"x\":3,}", + "{\"x\"}", + "{\"x\":3 q}", + "{\"x\":[3 4]}", + "{\"x\":[\"", + "{\"x\":[[], 4],\"y\",}", + "{\"x\":[3", + "{\"x\":[ null, false, true}", + }; + for(auto s:json_error_tests) + { + auto x = json::load(s); + if (x) + { + fail("should fail to parse ", s); + return; + } + } + } + + auto x = json::load(R"({"message":"hello, world"})"); + if (!x) + fail("fail to parse"); + ASSERT_EQUAL("hello, world", x["message"]); + ASSERT_EQUAL(1, x.size()); + ASSERT_EQUAL(false, x.has("mess")); + ASSERT_THROW(x["mess"]); + ASSERT_THROW(3 == x["message"]); + ASSERT_EQUAL(12, x["message"].size()); + + std::string s = R"({"int":3, "ints" :[1,2,3,4,5] })"; + auto y = json::load(s); + ASSERT_EQUAL(3, y["int"]); + ASSERT_EQUAL(3.0, y["int"]); + ASSERT_NOTEQUAL(3.01, y["int"]); + ASSERT_EQUAL(5, y["ints"].size()); + ASSERT_EQUAL(1, y["ints"][0]); + ASSERT_EQUAL(2, y["ints"][1]); + ASSERT_EQUAL(3, y["ints"][2]); + ASSERT_EQUAL(4, y["ints"][3]); + ASSERT_EQUAL(5, y["ints"][4]); + ASSERT_EQUAL(1u, y["ints"][0]); + ASSERT_EQUAL(1.f, y["ints"][0]); + + int q = (int)y["ints"][1]; + ASSERT_EQUAL(2, q); + q = y["ints"][2].i(); + ASSERT_EQUAL(3, q); + +} + +TEST(json_read_unescaping) +{ + { + auto x = json::load(R"({"data":"\ud55c\n\t\r"})"); + if (!x) + { + fail("fail to parse"); + return; + } + ASSERT_EQUAL(6, x["data"].size()); + ASSERT_EQUAL("한\n\t\r", x["data"]); + } + { + // multiple r_string instance + auto x = json::load(R"({"data":"\ud55c\n\t\r"})"); + auto a = x["data"].s(); + auto b = x["data"].s(); + ASSERT_EQUAL(6, a.size()); + ASSERT_EQUAL(6, b.size()); + ASSERT_EQUAL(6, x["data"].size()); + } +} + +TEST(json_write) +{ + json::wvalue x; + x["message"] = "hello world"; + ASSERT_EQUAL(R"({"message":"hello world"})", json::dump(x)); + x["message"] = std::string("string value"); + ASSERT_EQUAL(R"({"message":"string value"})", json::dump(x)); + x["message"]["x"] = 3; + ASSERT_EQUAL(R"({"message":{"x":3}})", json::dump(x)); + x["message"]["y"] = 5; + ASSERT_TRUE(R"({"message":{"x":3,"y":5}})" == json::dump(x) || R"({"message":{"y":5,"x":3}})" == json::dump(x)); + x["message"] = 5.5; + ASSERT_EQUAL(R"({"message":5.5})", json::dump(x)); + + json::wvalue y; + y["scores"][0] = 1; + y["scores"][1] = "king"; + y["scores"][2] = 3.5; + ASSERT_EQUAL(R"({"scores":[1,"king",3.5]})", json::dump(y)); + + y["scores"][2][0] = "real"; + y["scores"][2][1] = false; + y["scores"][2][2] = true; + ASSERT_EQUAL(R"({"scores":[1,"king",["real",false,true]]})", json::dump(y)); + + y["scores"]["a"]["b"]["c"] = nullptr; + ASSERT_EQUAL(R"({"scores":{"a":{"b":{"c":null}}}})", json::dump(y)); +} + +TEST(template_basic) +{ + auto t = crow::mustache::compile(R"---(attack of {{name}})---"); + crow::mustache::context ctx; + ctx["name"] = "killer tomatoes"; + auto result = t.render(ctx); + ASSERT_EQUAL("attack of killer tomatoes", result); + //crow::mustache::load("basic.mustache"); +} + +TEST(template_load) +{ + crow::mustache::set_base("."); + ofstream("test.mustache") << R"---(attack of {{name}})---"; + auto t = crow::mustache::load("test.mustache"); + crow::mustache::context ctx; + ctx["name"] = "killer tomatoes"; + auto result = t.render(ctx); + ASSERT_EQUAL("attack of killer tomatoes", result); + unlink("test.mustache"); +} + +int testmain() +{ + bool failed = false; + for(auto t:tests) + { + failed__ = false; + try + { + //cerr << typeid(*t).name() << endl; + t->test(); + } + catch(std::exception& e) + { + fail(e.what()); + } + if (failed__) + { + cerr << "F"; + cerr << '\t' << typeid(*t).name() << endl; + failed = true; + } + else + cerr << "."; + } + cerr< Date: Thu, 7 Aug 2014 04:46:28 +0900 Subject: cmake update, added template test --- .gitignore | 3 ++- CMakeLists.txt | 14 +++++++------- examples/CMakeLists.txt | 11 ++++++----- examples/example_test.py | 28 ++++++++++++++++++++++++++++ src/CMakeLists.txt | 1 - tests/CMakeLists.txt | 23 ++++++++++++++--------- tests/template/CMakeLists.txt | 29 +++++++++++++++++++++++++++++ tests/template/mustachetest.cc | 33 --------------------------------- tests/template/mustachetest.cpp | 33 +++++++++++++++++++++++++++++++++ tests/template/test.py | 15 ++++++++------- tests/test.py | 28 ---------------------------- 11 files changed, 127 insertions(+), 91 deletions(-) create mode 100644 examples/example_test.py create mode 100644 tests/template/CMakeLists.txt delete mode 100644 tests/template/mustachetest.cc create mode 100644 tests/template/mustachetest.cpp delete mode 100644 tests/test.py (limited to 'tests') diff --git a/.gitignore b/.gitignore index ec63977..f3e7b9f 100644 --- a/.gitignore +++ b/.gitignore @@ -26,8 +26,9 @@ unittest *.swp *.gcov -covtest *.gcda *.gcno +build + .directory diff --git a/CMakeLists.txt b/CMakeLists.txt index 9fd219f..b517ae4 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -1,14 +1,10 @@ cmake_minimum_required(VERSION 2.8) project (crow_all) - -find_package( Boost 1.40 COMPONENTS date_time filesystem system thread REQUIRED ) -#find_package( Tcmalloc ) -include_directories( ${Boost_INCLUDE_DIR} ) -#set(CMAKE_BINARY_DIR ${CMAKE_SOURCE_DIR}/build) +set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -std=c++1y") -#set(EXECUTABLE_OUTPUT_PATH ${CMAKE_BINARY_DIR}) -#set(LIBRARY_OUTPUT_PATH ${CMAKE_BINARY_DIR}) +find_package( Boost 1.40 COMPONENTS date_time filesystem system thread REQUIRED ) +include_directories( ${Boost_INCLUDE_DIR} ) set(PROJECT_INCLUDE_DIR ${PROJECT_SOURCE_DIR}/include @@ -21,3 +17,7 @@ include_directories("${PROJECT_SOURCE_DIR}") add_subdirectory(src) add_subdirectory(tests) add_subdirectory(examples) + +enable_testing() +add_test(NAME crow_test COMMAND ${CMAKE_CURRENT_BINARY_DIR}/tests/unittest) +add_test(NAME template_test COMMAND ${CMAKE_CURRENT_BINARY_DIR}/tests/template/test.py WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/tests/template) diff --git a/examples/CMakeLists.txt b/examples/CMakeLists.txt index 7313adf..24150f0 100644 --- a/examples/CMakeLists.txt +++ b/examples/CMakeLists.txt @@ -4,16 +4,17 @@ project (crow_examples) add_executable(example example.cpp) target_link_libraries(example crow) target_link_libraries(example ${Boost_LIBRARIES} ) -set_target_properties(example PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") +add_custom_command(OUTPUT example_test.py + COMMAND ${CMAKE_COMMAND} -E + copy ${PROJECT_SOURCE_DIR}/example_test.py ${CMAKE_CURRENT_BINARY_DIR}/example_test.py + DEPENDS ${PROJECT_SOURCE_DIR}/example_test.py + ) +add_custom_target(example_copy ALL DEPENDS example_test.py) add_executable(example_chat example_chat.cpp) target_link_libraries(example_chat crow) target_link_libraries(example_chat ${Boost_LIBRARIES} ) -set_target_properties(example_chat PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") -message(${CMAKE_CURRENT_BINARY_DIR}) -message(${PROJECT_SOURCE_DIR}) add_custom_command(OUTPUT example_chat.html - #TARGET example_chat COMMAND ${CMAKE_COMMAND} -E copy ${PROJECT_SOURCE_DIR}/example_chat.html ${CMAKE_CURRENT_BINARY_DIR}/example_chat.html DEPENDS ${PROJECT_SOURCE_DIR}/example_chat.html diff --git a/examples/example_test.py b/examples/example_test.py new file mode 100644 index 0000000..e2bf651 --- /dev/null +++ b/examples/example_test.py @@ -0,0 +1,28 @@ +import urllib +assert "Hello World!" == urllib.urlopen('http://localhost:18080').read() +assert "About Crow example." == urllib.urlopen('http://localhost:18080/about').read() +assert 404 == urllib.urlopen('http://localhost:18080/list').getcode() +assert "3 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/3').read() +assert "100 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/100').read() +assert 400 == urllib.urlopen('http://localhost:18080/hello/500').getcode() +assert "3" == urllib.urlopen('http://localhost:18080/add_json', data='{"a":1,"b":2}').read() +assert "3" == urllib.urlopen('http://localhost:18080/add/1/2').read() + +# test persistent connection +import socket +import time +s = socket.socket() +s.connect(('localhost', 18080)) +for i in xrange(10): + s.send('''GET / HTTP/1.1 +Host: localhost\r\n\r\n'''); + assert 'Hello World!' in s.recv(1024) + +# test timeout +s = socket.socket() +s.connect(('localhost', 18080)) +print 'ERROR REQUEST' +s.send('''GET / HTTP/1.1 +hHhHHefhwjkefhklwejfklwejf +''') +print s.recv(1024) diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index bf74895..4ac6f63 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -15,4 +15,3 @@ include_directories("${PROJECT_INCLUDE_DIR}") add_library(${PROJECT_NAME} SHARED ${CROW_SRCS}) #target_link_libraries(${PROJECT_NAME} tcmalloc) -set_target_properties(${PROJECT_NAME} PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt index 45943ed..d35925c 100644 --- a/tests/CMakeLists.txt +++ b/tests/CMakeLists.txt @@ -2,16 +2,21 @@ cmake_minimum_required(VERSION 2.8) project (crow_test) -set(PROJECT_INCLUDE_DIR -${PROJECT_SOURCE_DIR}/include -${PROJECT_SOURCE_DIR}/http-parser -) - set(TEST_SRCS unittest.cpp ) -add_executable(test ${TEST_SRCS}) -target_link_libraries(test crow) -target_link_libraries( test ${Boost_LIBRARIES} ) -set_target_properties(test PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") +add_executable(unittest ${TEST_SRCS}) +target_link_libraries(unittest crow) +target_link_libraries(unittest ${Boost_LIBRARIES} ) + +set_target_properties(unittest PROPERTIES COMPILE_FLAGS "--coverage -fprofile-arcs -ftest-coverage") + +target_link_libraries(unittest gcov) + +add_subdirectory(template) +#CXXFLAGS="-g -O0 -Wall -W -Wshadow -Wunused-variable \ +#Wunused-parameter -Wunused-function -Wunused -Wno-system-headers \ +#-Wno-deprecated -Woverloaded-virtual -Wwrite-strings -fprofile-arcs -ftest-coverage" +#CFLAGS="-g -O0 -Wall -W -fprofile-arcs -ftest-coverage" +#LDFLAGS="-fprofile-arcs -ftest-coverage" diff --git a/tests/template/CMakeLists.txt b/tests/template/CMakeLists.txt new file mode 100644 index 0000000..1d4ec30 --- /dev/null +++ b/tests/template/CMakeLists.txt @@ -0,0 +1,29 @@ +cmake_minimum_required(VERSION 2.8) +project (template_test) + + +set(PROJECT_INCLUDE_DIR +${PROJECT_SOURCE_DIR}/include +) + +set(TEST_SRCS +mustachetest.cpp +) + +add_executable(mustachetest ${TEST_SRCS}) +#target_link_libraries(unittest crow) +#target_link_libraries(unittest ${Boost_LIBRARIES} ) +set_target_properties(mustachetest PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") + +message(${PROJECT_SOURCE_DIR}) +message(${CMAKE_CURRENT_BINARY_DIR}) +file(COPY DIRECTORY . DESTINATION ${CMAKE_CURRENT_BINARY_DIR} + FILES_MATCHING + PATTERN "*.json") + +add_custom_command(OUTPUT test.py + COMMAND ${CMAKE_COMMAND} -E + copy ${PROJECT_SOURCE_DIR}/test.py ${CMAKE_CURRENT_BINARY_DIR}/test.py + DEPENDS ${PROJECT_SOURCE_DIR}/test.py + ) +add_custom_target(template_test_copy ALL DEPENDS test.py) diff --git a/tests/template/mustachetest.cc b/tests/template/mustachetest.cc deleted file mode 100644 index c4ac2c9..0000000 --- a/tests/template/mustachetest.cc +++ /dev/null @@ -1,33 +0,0 @@ -#include -#include -#include -#include -#include "../mustache.h" -#include "../json.h" -using namespace std; -using namespace crow; -using namespace crow::mustache; - -string read_all(const string& filename) -{ - ifstream is(filename); - return {istreambuf_iterator(is), istreambuf_iterator()}; -} - -int main() -{ - auto data = json::load(read_all("data")); - auto templ = compile(read_all("template")); - auto partials = json::load(read_all("partials")); - set_loader([&](std::string name)->std::string - { - if (partials.count(name)) - { - return partials[name].s(); - } - return ""; - }); - context ctx(data); - cout << templ.render(ctx); - return 0; -} diff --git a/tests/template/mustachetest.cpp b/tests/template/mustachetest.cpp new file mode 100644 index 0000000..1caf537 --- /dev/null +++ b/tests/template/mustachetest.cpp @@ -0,0 +1,33 @@ +#include +#include +#include +#include +#include "mustache.h" +#include "json.h" +using namespace std; +using namespace crow; +using namespace crow::mustache; + +string read_all(const string& filename) +{ + ifstream is(filename); + return {istreambuf_iterator(is), istreambuf_iterator()}; +} + +int main() +{ + auto data = json::load(read_all("data")); + auto templ = compile(read_all("template")); + auto partials = json::load(read_all("partials")); + set_loader([&](std::string name)->std::string + { + if (partials.count(name)) + { + return partials[name].s(); + } + return ""; + }); + context ctx(data); + cout << templ.render(ctx); + return 0; +} diff --git a/tests/template/test.py b/tests/template/test.py index 22fcca0..57bcaf8 100755 --- a/tests/template/test.py +++ b/tests/template/test.py @@ -1,4 +1,5 @@ #!/usr/bin/env python +from __future__ import print_function import glob import json import os @@ -14,15 +15,15 @@ for testfile in glob.glob("*.json"): open('partials', 'w').write(json.dumps(test["partials"])) else: open('partials', 'w').write("{}") - ret = subprocess.check_output("./mustachetest") - print testfile, test["name"] + ret = subprocess.check_output("./mustachetest").decode('utf8') + print(testfile, test["name"]) if ret != test["expected"]: if 'partials' in test: - print 'partials:', json.dumps(test["partials"]) - print json.dumps(test["data"]) - print test["template"] - print 'Expected:',repr(test["expected"]) - print 'Actual:',repr(ret) + print('partials:', json.dumps(test["partials"])) + print(json.dumps(test["data"])) + print(test["template"]) + print('Expected:',repr(test["expected"])) + print('Actual:',repr(ret)) assert ret == test["expected"] os.unlink('data') os.unlink('template') diff --git a/tests/test.py b/tests/test.py deleted file mode 100644 index e2bf651..0000000 --- a/tests/test.py +++ /dev/null @@ -1,28 +0,0 @@ -import urllib -assert "Hello World!" == urllib.urlopen('http://localhost:18080').read() -assert "About Crow example." == urllib.urlopen('http://localhost:18080/about').read() -assert 404 == urllib.urlopen('http://localhost:18080/list').getcode() -assert "3 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/3').read() -assert "100 bottles of beer!" == urllib.urlopen('http://localhost:18080/hello/100').read() -assert 400 == urllib.urlopen('http://localhost:18080/hello/500').getcode() -assert "3" == urllib.urlopen('http://localhost:18080/add_json', data='{"a":1,"b":2}').read() -assert "3" == urllib.urlopen('http://localhost:18080/add/1/2').read() - -# test persistent connection -import socket -import time -s = socket.socket() -s.connect(('localhost', 18080)) -for i in xrange(10): - s.send('''GET / HTTP/1.1 -Host: localhost\r\n\r\n'''); - assert 'Hello World!' in s.recv(1024) - -# test timeout -s = socket.socket() -s.connect(('localhost', 18080)) -print 'ERROR REQUEST' -s.send('''GET / HTTP/1.1 -hHhHHefhwjkefhklwejfklwejf -''') -print s.recv(1024) -- cgit v1.2.3-54-g00ecf From c36aa219e25e26019035fe3465471fd510778422 Mon Sep 17 00:00:00 2001 From: ipknHama Date: Thu, 7 Aug 2014 08:31:27 +0900 Subject: now crow becomes completely header only --- CMakeLists.txt | 7 +- examples/CMakeLists.txt | 4 +- http-parser | 2 +- include/http_connection.h | 9 +- include/http_parser_merged.h | 2640 +++++++++++++++++++++++++++++++++++++++++ include/mustache.h | 2 +- src/CMakeLists.txt | 17 - tests/CMakeLists.txt | 2 +- tests/template/CMakeLists.txt | 4 +- 9 files changed, 2659 insertions(+), 28 deletions(-) create mode 100644 include/http_parser_merged.h delete mode 100644 src/CMakeLists.txt (limited to 'tests') diff --git a/CMakeLists.txt b/CMakeLists.txt index b517ae4..338e740 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -1,6 +1,11 @@ cmake_minimum_required(VERSION 2.8) project (crow_all) +if (NOT CMAKE_BUILD_TYPE) + message(STATUS "No build type selected, default to Release") + set(CMAKE_BUILD_TYPE "Release") +endif() + set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -std=c++1y") find_package( Boost 1.40 COMPONENTS date_time filesystem system thread REQUIRED ) @@ -14,7 +19,7 @@ ${PROJECT_SOURCE_DIR}/http-parser include_directories("${PROJECT_INCLUDE_DIR}") include_directories("${PROJECT_SOURCE_DIR}") -add_subdirectory(src) +#add_subdirectory(src) add_subdirectory(tests) add_subdirectory(examples) diff --git a/examples/CMakeLists.txt b/examples/CMakeLists.txt index 24150f0..031cc04 100644 --- a/examples/CMakeLists.txt +++ b/examples/CMakeLists.txt @@ -2,7 +2,7 @@ cmake_minimum_required(VERSION 2.8) project (crow_examples) add_executable(example example.cpp) -target_link_libraries(example crow) +#target_link_libraries(example crow) target_link_libraries(example ${Boost_LIBRARIES} ) add_custom_command(OUTPUT example_test.py COMMAND ${CMAKE_COMMAND} -E @@ -12,7 +12,7 @@ add_custom_command(OUTPUT example_test.py add_custom_target(example_copy ALL DEPENDS example_test.py) add_executable(example_chat example_chat.cpp) -target_link_libraries(example_chat crow) +#target_link_libraries(example_chat crow) target_link_libraries(example_chat ${Boost_LIBRARIES} ) add_custom_command(OUTPUT example_chat.html COMMAND ${CMAKE_COMMAND} -E diff --git a/http-parser b/http-parser index d19e129..5b951d7 160000 --- a/http-parser +++ b/http-parser @@ -1 +1 @@ -Subproject commit d19e12911a1b523885bdf0d82f1f3e91df90711d +Subproject commit 5b951d74bd66ec9d38448e0a85b1cf8b85d97db3 diff --git a/include/http_connection.h b/include/http_connection.h index d73398e..a030d73 100644 --- a/include/http_connection.h +++ b/include/http_connection.h @@ -1,11 +1,12 @@ #pragma once #include #include +#include #include #include #include -#include +#include "http_parser_merged.h" #include "datetime.h" #include "parser.h" @@ -29,7 +30,8 @@ namespace crow handler_(handler), parser_(this), server_name_(server_name), - deadline_(socket_.get_io_service()) + deadline_(socket_.get_io_service()), + address_str_(boost::lexical_cast(socket_.remote_endpoint())) { #ifdef CROW_ENABLE_DEBUG connectionCount ++; @@ -89,7 +91,7 @@ namespace crow } } - CROW_LOG_INFO << "Request: "<< this << " HTTP/" << parser_.http_major << "." << parser_.http_minor << ' ' + CROW_LOG_INFO << "Request: " << address_str_ << " " << this << " HTTP/" << parser_.http_major << "." << parser_.http_minor << ' ' << method_name(req.method) << " " << req.url; @@ -308,6 +310,7 @@ namespace crow std::string date_str_; boost::asio::deadline_timer deadline_; + std::string address_str_; }; } diff --git a/include/http_parser_merged.h b/include/http_parser_merged.h new file mode 100644 index 0000000..c672352 --- /dev/null +++ b/include/http_parser_merged.h @@ -0,0 +1,2640 @@ +/* merged revision: 5b951d74bd66ec9d38448e0a85b1cf8b85d97db3 */ +/* Copyright Joyent, Inc. and other Node contributors. All rights reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ +#ifndef http_parser_h +#define http_parser_h +#ifdef __cplusplus +extern "C" { +#endif + +/* Also update SONAME in the Makefile whenever you change these. */ +#define HTTP_PARSER_VERSION_MAJOR 2 +#define HTTP_PARSER_VERSION_MINOR 3 +#define HTTP_PARSER_VERSION_PATCH 0 + +#include +#if defined(_WIN32) && !defined(__MINGW32__) && (!defined(_MSC_VER) || _MSC_VER<1600) +#include +#include +typedef __int8 int8_t; +typedef unsigned __int8 uint8_t; +typedef __int16 int16_t; +typedef unsigned __int16 uint16_t; +typedef __int32 int32_t; +typedef unsigned __int32 uint32_t; +typedef __int64 int64_t; +typedef unsigned __int64 uint64_t; +#else +#include +#endif + +/* Compile with -DHTTP_PARSER_STRICT=0 to make less checks, but run + * faster + */ +#ifndef HTTP_PARSER_STRICT +# define HTTP_PARSER_STRICT 1 +#endif + +/* Maximium header size allowed. If the macro is not defined + * before including this header then the default is used. To + * change the maximum header size, define the macro in the build + * environment (e.g. -DHTTP_MAX_HEADER_SIZE=). To remove + * the effective limit on the size of the header, define the macro + * to a very large number (e.g. -DHTTP_MAX_HEADER_SIZE=0x7fffffff) + */ +#ifndef HTTP_MAX_HEADER_SIZE +# define HTTP_MAX_HEADER_SIZE (80*1024) +#endif + +typedef struct http_parser http_parser; +typedef struct http_parser_settings http_parser_settings; + + +/* Callbacks should return non-zero to indicate an error. The parser will + * then halt execution. + * + * The one exception is on_headers_complete. In a HTTP_RESPONSE parser + * returning '1' from on_headers_complete will tell the parser that it + * should not expect a body. This is used when receiving a response to a + * HEAD request which may contain 'Content-Length' or 'Transfer-Encoding: + * chunked' headers that indicate the presence of a body. + * + * http_data_cb does not return data chunks. It will be call arbitrarally + * many times for each string. E.G. you might get 10 callbacks for "on_url" + * each providing just a few characters more data. + */ +typedef int (*http_data_cb) (http_parser*, const char *at, size_t length); +typedef int (*http_cb) (http_parser*); + + +/* Request Methods */ +#define HTTP_METHOD_MAP(XX) \ + XX(0, DELETE, DELETE) \ + XX(1, GET, GET) \ + XX(2, HEAD, HEAD) \ + XX(3, POST, POST) \ + XX(4, PUT, PUT) \ + /* pathological */ \ + XX(5, CONNECT, CONNECT) \ + XX(6, OPTIONS, OPTIONS) \ + XX(7, TRACE, TRACE) \ + /* webdav */ \ + XX(8, COPY, COPY) \ + XX(9, LOCK, LOCK) \ + XX(10, MKCOL, MKCOL) \ + XX(11, MOVE, MOVE) \ + XX(12, PROPFIND, PROPFIND) \ + XX(13, PROPPATCH, PROPPATCH) \ + XX(14, SEARCH, SEARCH) \ + XX(15, UNLOCK, UNLOCK) \ + /* subversion */ \ + XX(16, REPORT, REPORT) \ + XX(17, MKACTIVITY, MKACTIVITY) \ + XX(18, CHECKOUT, CHECKOUT) \ + XX(19, MERGE, MERGE) \ + /* upnp */ \ + XX(20, MSEARCH, M-SEARCH) \ + XX(21, NOTIFY, NOTIFY) \ + XX(22, SUBSCRIBE, SUBSCRIBE) \ + XX(23, UNSUBSCRIBE, UNSUBSCRIBE) \ + /* RFC-5789 */ \ + XX(24, PATCH, PATCH) \ + XX(25, PURGE, PURGE) \ + /* CalDAV */ \ + XX(26, MKCALENDAR, MKCALENDAR) \ + +enum http_method + { +#define XX(num, name, string) HTTP_##name = num, + HTTP_METHOD_MAP(XX) +#undef XX + }; + + +enum http_parser_type { HTTP_REQUEST, HTTP_RESPONSE, HTTP_BOTH }; + + +/* Flag values for http_parser.flags field */ +enum flags + { F_CHUNKED = 1 << 0 + , F_CONNECTION_KEEP_ALIVE = 1 << 1 + , F_CONNECTION_CLOSE = 1 << 2 + , F_TRAILING = 1 << 3 + , F_UPGRADE = 1 << 4 + , F_SKIPBODY = 1 << 5 + }; + + +/* Map for errno-related constants + * + * The provided argument should be a macro that takes 2 arguments. + */ +#define HTTP_ERRNO_MAP(XX) \ + /* No error */ \ + XX(OK, "success") \ + \ + /* Callback-related errors */ \ + XX(CB_message_begin, "the on_message_begin callback failed") \ + XX(CB_url, "the on_url callback failed") \ + XX(CB_header_field, "the on_header_field callback failed") \ + XX(CB_header_value, "the on_header_value callback failed") \ + XX(CB_headers_complete, "the on_headers_complete callback failed") \ + XX(CB_body, "the on_body callback failed") \ + XX(CB_message_complete, "the on_message_complete callback failed") \ + XX(CB_status, "the on_status callback failed") \ + \ + /* Parsing-related errors */ \ + XX(INVALID_EOF_STATE, "stream ended at an unexpected time") \ + XX(HEADER_OVERFLOW, \ + "too many header bytes seen; overflow detected") \ + XX(CLOSED_CONNECTION, \ + "data received after completed connection: close message") \ + XX(INVALID_VERSION, "invalid HTTP version") \ + XX(INVALID_STATUS, "invalid HTTP status code") \ + XX(INVALID_METHOD, "invalid HTTP method") \ + XX(INVALID_URL, "invalid URL") \ + XX(INVALID_HOST, "invalid host") \ + XX(INVALID_PORT, "invalid port") \ + XX(INVALID_PATH, "invalid path") \ + XX(INVALID_QUERY_STRING, "invalid query string") \ + XX(INVALID_FRAGMENT, "invalid fragment") \ + XX(LF_EXPECTED, "LF character expected") \ + XX(INVALID_HEADER_TOKEN, "invalid character in header") \ + XX(INVALID_CONTENT_LENGTH, \ + "invalid character in content-length header") \ + XX(INVALID_CHUNK_SIZE, \ + "invalid character in chunk size header") \ + XX(INVALID_CONSTANT, "invalid constant string") \ + XX(INVALID_INTERNAL_STATE, "encountered unexpected internal state")\ + XX(STRICT, "strict mode assertion failed") \ + XX(PAUSED, "parser is paused") \ + XX(UNKNOWN, "an unknown error occurred") + + +/* Define HPE_* values for each errno value above */ +#define HTTP_ERRNO_GEN(n, s) HPE_##n, +enum http_errno { + HTTP_ERRNO_MAP(HTTP_ERRNO_GEN) +}; +#undef HTTP_ERRNO_GEN + + +/* Get an http_errno value from an http_parser */ +#define HTTP_PARSER_ERRNO(p) ((enum http_errno) (p)->http_errno) + + +struct http_parser { + /** PRIVATE **/ + unsigned int type : 2; /* enum http_parser_type */ + unsigned int flags : 6; /* F_* values from 'flags' enum; semi-public */ + unsigned int state : 8; /* enum state from http_parser.c */ + unsigned int header_state : 8; /* enum header_state from http_parser.c */ + unsigned int index : 8; /* index into current matcher */ + + uint32_t nread; /* # bytes read in various scenarios */ + uint64_t content_length; /* # bytes in body (0 if no Content-Length header) */ + + /** READ-ONLY **/ + unsigned short http_major; + unsigned short http_minor; + unsigned int status_code : 16; /* responses only */ + unsigned int method : 8; /* requests only */ + unsigned int http_errno : 7; + + /* 1 = Upgrade header was present and the parser has exited because of that. + * 0 = No upgrade header present. + * Should be checked when http_parser_execute() returns in addition to + * error checking. + */ + unsigned int upgrade : 1; + + /** PUBLIC **/ + void *data; /* A pointer to get hook to the "connection" or "socket" object */ +}; + + +struct http_parser_settings { + http_cb on_message_begin; + http_data_cb on_url; + http_data_cb on_status; + http_data_cb on_header_field; + http_data_cb on_header_value; + http_cb on_headers_complete; + http_data_cb on_body; + http_cb on_message_complete; +}; + + +enum http_parser_url_fields + { UF_SCHEMA = 0 + , UF_HOST = 1 + , UF_PORT = 2 + , UF_PATH = 3 + , UF_QUERY = 4 + , UF_FRAGMENT = 5 + , UF_USERINFO = 6 + , UF_MAX = 7 + }; + + +/* Result structure for http_parser_parse_url(). + * + * Callers should index into field_data[] with UF_* values iff field_set + * has the relevant (1 << UF_*) bit set. As a courtesy to clients (and + * because we probably have padding left over), we convert any port to + * a uint16_t. + */ +struct http_parser_url { + uint16_t field_set; /* Bitmask of (1 << UF_*) values */ + uint16_t port; /* Converted UF_PORT string */ + + struct { + uint16_t off; /* Offset into buffer in which field starts */ + uint16_t len; /* Length of run in buffer */ + } field_data[UF_MAX]; +}; + + +/* Returns the library version. Bits 16-23 contain the major version number, + * bits 8-15 the minor version number and bits 0-7 the patch level. + * Usage example: + * + * unsigned long version = http_parser_version(); + * unsigned major = (version >> 16) & 255; + * unsigned minor = (version >> 8) & 255; + * unsigned patch = version & 255; + * printf("http_parser v%u.%u.%u\n", major, minor, version); + */ +unsigned long http_parser_version(void); + +void http_parser_init(http_parser *parser, enum http_parser_type type); + + +size_t http_parser_execute(http_parser *parser, + const http_parser_settings *settings, + const char *data, + size_t len); + + +/* If http_should_keep_alive() in the on_headers_complete or + * on_message_complete callback returns 0, then this should be + * the last message on the connection. + * If you are the server, respond with the "Connection: close" header. + * If you are the client, close the connection. + */ +int http_should_keep_alive(const http_parser *parser); + +/* Returns a string version of the HTTP method. */ +const char *http_method_str(enum http_method m); + +/* Return a string name of the given error */ +const char *http_errno_name(enum http_errno err); + +/* Return a string description of the given error */ +const char *http_errno_description(enum http_errno err); + +/* Parse a URL; return nonzero on failure */ +int http_parser_parse_url(const char *buf, size_t buflen, + int is_connect, + struct http_parser_url *u); + +/* Pause or un-pause the parser; a nonzero value pauses */ +void http_parser_pause(http_parser *parser, int paused); + +/* Checks if this is the final chunk of the body. */ +int http_body_is_final(const http_parser *parser); + +/*#include "http_parser.h"*/ +/* Based on src/http/ngx_http_parse.c from NGINX copyright Igor Sysoev + * + * Additional changes are licensed under the same terms as NGINX and + * copyright Joyent, Inc. and other Node contributors. All rights reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ +#include +#include +#include +#include +#include +#include + +#ifndef ULLONG_MAX +# define ULLONG_MAX ((uint64_t) -1) /* 2^64-1 */ +#endif + +#ifndef MIN +# define MIN(a,b) ((a) < (b) ? (a) : (b)) +#endif + +#ifndef ARRAY_SIZE +# define ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0])) +#endif + +#ifndef BIT_AT +# define BIT_AT(a, i) \ + (!!((unsigned int) (a)[(unsigned int) (i) >> 3] & \ + (1 << ((unsigned int) (i) & 7)))) +#endif + +#ifndef ELEM_AT +# define ELEM_AT(a, i, v) ((unsigned int) (i) < ARRAY_SIZE(a) ? (a)[(i)] : (v)) +#endif + +#define SET_ERRNO(e) \ +do { \ + parser->http_errno = (e); \ +} while(0) + + +/* Run the notify callback FOR, returning ER if it fails */ +#define CALLBACK_NOTIFY_(FOR, ER) \ +do { \ + assert(HTTP_PARSER_ERRNO(parser) == HPE_OK); \ + \ + if (settings->on_##FOR) { \ + if (0 != settings->on_##FOR(parser)) { \ + SET_ERRNO(HPE_CB_##FOR); \ + } \ + \ + /* We either errored above or got paused; get out */ \ + if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { \ + return (ER); \ + } \ + } \ +} while (0) + +/* Run the notify callback FOR and consume the current byte */ +#define CALLBACK_NOTIFY(FOR) CALLBACK_NOTIFY_(FOR, p - data + 1) + +/* Run the notify callback FOR and don't consume the current byte */ +#define CALLBACK_NOTIFY_NOADVANCE(FOR) CALLBACK_NOTIFY_(FOR, p - data) + +/* Run data callback FOR with LEN bytes, returning ER if it fails */ +#define CALLBACK_DATA_(FOR, LEN, ER) \ +do { \ + assert(HTTP_PARSER_ERRNO(parser) == HPE_OK); \ + \ + if (FOR##_mark) { \ + if (settings->on_##FOR) { \ + if (0 != settings->on_##FOR(parser, FOR##_mark, (LEN))) { \ + SET_ERRNO(HPE_CB_##FOR); \ + } \ + \ + /* We either errored above or got paused; get out */ \ + if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { \ + return (ER); \ + } \ + } \ + FOR##_mark = NULL; \ + } \ +} while (0) + +/* Run the data callback FOR and consume the current byte */ +#define CALLBACK_DATA(FOR) \ + CALLBACK_DATA_(FOR, p - FOR##_mark, p - data + 1) + +/* Run the data callback FOR and don't consume the current byte */ +#define CALLBACK_DATA_NOADVANCE(FOR) \ + CALLBACK_DATA_(FOR, p - FOR##_mark, p - data) + +/* Set the mark FOR; non-destructive if mark is already set */ +#define MARK(FOR) \ +do { \ + if (!FOR##_mark) { \ + FOR##_mark = p; \ + } \ +} while (0) + + +#define PROXY_CONNECTION "proxy-connection" +#define CONNECTION "connection" +#define CONTENT_LENGTH "content-length" +#define TRANSFER_ENCODING "transfer-encoding" +#define UPGRADE "upgrade" +#define CHUNKED "chunked" +#define KEEP_ALIVE "keep-alive" +#define CLOSE "close" + + + + +enum state + { s_dead = 1 /* important that this is > 0 */ + + , s_start_req_or_res + , s_res_or_resp_H + , s_start_res + , s_res_H + , s_res_HT + , s_res_HTT + , s_res_HTTP + , s_res_first_http_major + , s_res_http_major + , s_res_first_http_minor + , s_res_http_minor + , s_res_first_status_code + , s_res_status_code + , s_res_status_start + , s_res_status + , s_res_line_almost_done + + , s_start_req + + , s_req_method + , s_req_spaces_before_url + , s_req_schema + , s_req_schema_slash + , s_req_schema_slash_slash + , s_req_server_start + , s_req_server + , s_req_server_with_at + , s_req_path + , s_req_query_string_start + , s_req_query_string + , s_req_fragment_start + , s_req_fragment + , s_req_http_start + , s_req_http_H + , s_req_http_HT + , s_req_http_HTT + , s_req_http_HTTP + , s_req_first_http_major + , s_req_http_major + , s_req_first_http_minor + , s_req_http_minor + , s_req_line_almost_done + + , s_header_field_start + , s_header_field + , s_header_value_discard_ws + , s_header_value_discard_ws_almost_done + , s_header_value_discard_lws + , s_header_value_start + , s_header_value + , s_header_value_lws + + , s_header_almost_done + + , s_chunk_size_start + , s_chunk_size + , s_chunk_parameters + , s_chunk_size_almost_done + + , s_headers_almost_done + , s_headers_done + + /* Important: 's_headers_done' must be the last 'header' state. All + * states beyond this must be 'body' states. It is used for overflow + * checking. See the PARSING_HEADER() macro. + */ + + , s_chunk_data + , s_chunk_data_almost_done + , s_chunk_data_done + + , s_body_identity + , s_body_identity_eof + + , s_message_done + }; + + +#define PARSING_HEADER(state) (state <= s_headers_done) + + +enum header_states + { h_general = 0 + , h_C + , h_CO + , h_CON + + , h_matching_connection + , h_matching_proxy_connection + , h_matching_content_length + , h_matching_transfer_encoding + , h_matching_upgrade + + , h_connection + , h_content_length + , h_transfer_encoding + , h_upgrade + + , h_matching_transfer_encoding_chunked + , h_matching_connection_keep_alive + , h_matching_connection_close + + , h_transfer_encoding_chunked + , h_connection_keep_alive + , h_connection_close + }; + +enum http_host_state + { + s_http_host_dead = 1 + , s_http_userinfo_start + , s_http_userinfo + , s_http_host_start + , s_http_host_v6_start + , s_http_host + , s_http_host_v6 + , s_http_host_v6_end + , s_http_host_port_start + , s_http_host_port +}; + +/* Macros for character classes; depends on strict-mode */ +#define CR '\r' +#define LF '\n' +#define LOWER(c) (unsigned char)(c | 0x20) +#define IS_ALPHA(c) (LOWER(c) >= 'a' && LOWER(c) <= 'z') +#define IS_NUM(c) ((c) >= '0' && (c) <= '9') +#define IS_ALPHANUM(c) (IS_ALPHA(c) || IS_NUM(c)) +#define IS_HEX(c) (IS_NUM(c) || (LOWER(c) >= 'a' && LOWER(c) <= 'f')) +#define IS_MARK(c) ((c) == '-' || (c) == '_' || (c) == '.' || \ + (c) == '!' || (c) == '~' || (c) == '*' || (c) == '\'' || (c) == '(' || \ + (c) == ')') +#define IS_USERINFO_CHAR(c) (IS_ALPHANUM(c) || IS_MARK(c) || (c) == '%' || \ + (c) == ';' || (c) == ':' || (c) == '&' || (c) == '=' || (c) == '+' || \ + (c) == '$' || (c) == ',') + +#if HTTP_PARSER_STRICT +#define TOKEN(c) (tokens[(unsigned char)c]) +#define IS_URL_CHAR(c) (BIT_AT(normal_url_char, (unsigned char)c)) +#define IS_HOST_CHAR(c) (IS_ALPHANUM(c) || (c) == '.' || (c) == '-') +#else +#define TOKEN(c) ((c == ' ') ? ' ' : tokens[(unsigned char)c]) +#define IS_URL_CHAR(c) \ + (BIT_AT(normal_url_char, (unsigned char)c) || ((c) & 0x80)) +#define IS_HOST_CHAR(c) \ + (IS_ALPHANUM(c) || (c) == '.' || (c) == '-' || (c) == '_') +#endif + + +#define start_state (parser->type == HTTP_REQUEST ? s_start_req : s_start_res) + + +#if HTTP_PARSER_STRICT +# define STRICT_CHECK(cond) \ +do { \ + if (cond) { \ + SET_ERRNO(HPE_STRICT); \ + goto error; \ + } \ +} while (0) +# define NEW_MESSAGE() (http_should_keep_alive(parser) ? start_state : s_dead) +#else +# define STRICT_CHECK(cond) +# define NEW_MESSAGE() start_state +#endif + + + +int http_message_needs_eof(const http_parser *parser); + +/* Our URL parser. + * + * This is designed to be shared by http_parser_execute() for URL validation, + * hence it has a state transition + byte-for-byte interface. In addition, it + * is meant to be embedded in http_parser_parse_url(), which does the dirty + * work of turning state transitions URL components for its API. + * + * This function should only be invoked with non-space characters. It is + * assumed that the caller cares about (and can detect) the transition between + * URL and non-URL states by looking for these. + */ +enum state +parse_url_char(enum state s, const char ch) +{ +#if HTTP_PARSER_STRICT +# define T(v) 0 +#else +# define T(v) v +#endif + + +static const uint8_t normal_url_char[32] = { +/* 0 nul 1 soh 2 stx 3 etx 4 eot 5 enq 6 ack 7 bel */ + 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0, +/* 8 bs 9 ht 10 nl 11 vt 12 np 13 cr 14 so 15 si */ + 0 | T(2) | 0 | 0 | T(16) | 0 | 0 | 0, +/* 16 dle 17 dc1 18 dc2 19 dc3 20 dc4 21 nak 22 syn 23 etb */ + 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0, +/* 24 can 25 em 26 sub 27 esc 28 fs 29 gs 30 rs 31 us */ + 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0, +/* 32 sp 33 ! 34 " 35 # 36 $ 37 % 38 & 39 ' */ + 0 | 2 | 4 | 0 | 16 | 32 | 64 | 128, +/* 40 ( 41 ) 42 * 43 + 44 , 45 - 46 . 47 / */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 48 0 49 1 50 2 51 3 52 4 53 5 54 6 55 7 */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 56 8 57 9 58 : 59 ; 60 < 61 = 62 > 63 ? */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 0, +/* 64 @ 65 A 66 B 67 C 68 D 69 E 70 F 71 G */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 72 H 73 I 74 J 75 K 76 L 77 M 78 N 79 O */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 80 P 81 Q 82 R 83 S 84 T 85 U 86 V 87 W */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 88 X 89 Y 90 Z 91 [ 92 \ 93 ] 94 ^ 95 _ */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 96 ` 97 a 98 b 99 c 100 d 101 e 102 f 103 g */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 104 h 105 i 106 j 107 k 108 l 109 m 110 n 111 o */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 112 p 113 q 114 r 115 s 116 t 117 u 118 v 119 w */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128, +/* 120 x 121 y 122 z 123 { 124 | 125 } 126 ~ 127 del */ + 1 | 2 | 4 | 8 | 16 | 32 | 64 | 0, }; + +#undef T + + if (ch == ' ' || ch == '\r' || ch == '\n') { + return s_dead; + } + +#if HTTP_PARSER_STRICT + if (ch == '\t' || ch == '\f') { + return s_dead; + } +#endif + + switch (s) { + case s_req_spaces_before_url: + /* Proxied requests are followed by scheme of an absolute URI (alpha). + * All methods except CONNECT are followed by '/' or '*'. + */ + + if (ch == '/' || ch == '*') { + return s_req_path; + } + + if (IS_ALPHA(ch)) { + return s_req_schema; + } + + break; + + case s_req_schema: + if (IS_ALPHA(ch)) { + return s; + } + + if (ch == ':') { + return s_req_schema_slash; + } + + break; + + case s_req_schema_slash: + if (ch == '/') { + return s_req_schema_slash_slash; + } + + break; + + case s_req_schema_slash_slash: + if (ch == '/') { + return s_req_server_start; + } + + break; + + case s_req_server_with_at: + if (ch == '@') { + return s_dead; + } + + /* FALLTHROUGH */ + case s_req_server_start: + case s_req_server: + if (ch == '/') { + return s_req_path; + } + + if (ch == '?') { + return s_req_query_string_start; + } + + if (ch == '@') { + return s_req_server_with_at; + } + + if (IS_USERINFO_CHAR(ch) || ch == '[' || ch == ']') { + return s_req_server; + } + + break; + + case s_req_path: + if (IS_URL_CHAR(ch)) { + return s; + } + + switch (ch) { + case '?': + return s_req_query_string_start; + + case '#': + return s_req_fragment_start; + } + + break; + + case s_req_query_string_start: + case s_req_query_string: + if (IS_URL_CHAR(ch)) { + return s_req_query_string; + } + + switch (ch) { + case '?': + /* allow extra '?' in query string */ + return s_req_query_string; + + case '#': + return s_req_fragment_start; + } + + break; + + case s_req_fragment_start: + if (IS_URL_CHAR(ch)) { + return s_req_fragment; + } + + switch (ch) { + case '?': + return s_req_fragment; + + case '#': + return s; + } + + break; + + case s_req_fragment: + if (IS_URL_CHAR(ch)) { + return s; + } + + switch (ch) { + case '?': + case '#': + return s; + } + + break; + + default: + break; + } + + /* We should never fall out of the switch above unless there's an error */ + return s_dead; +} + +size_t http_parser_execute (http_parser *parser, + const http_parser_settings *settings, + const char *data, + size_t len) +{ +static const char *method_strings[] = + { +#define XX(num, name, string) #string, + HTTP_METHOD_MAP(XX) +#undef XX + }; + +/* Tokens as defined by rfc 2616. Also lowercases them. + * token = 1* + * separators = "(" | ")" | "<" | ">" | "@" + * | "," | ";" | ":" | "\" | <"> + * | "/" | "[" | "]" | "?" | "=" + * | "{" | "}" | SP | HT + */ +static const char tokens[256] = { +/* 0 nul 1 soh 2 stx 3 etx 4 eot 5 enq 6 ack 7 bel */ + 0, 0, 0, 0, 0, 0, 0, 0, +/* 8 bs 9 ht 10 nl 11 vt 12 np 13 cr 14 so 15 si */ + 0, 0, 0, 0, 0, 0, 0, 0, +/* 16 dle 17 dc1 18 dc2 19 dc3 20 dc4 21 nak 22 syn 23 etb */ + 0, 0, 0, 0, 0, 0, 0, 0, +/* 24 can 25 em 26 sub 27 esc 28 fs 29 gs 30 rs 31 us */ + 0, 0, 0, 0, 0, 0, 0, 0, +/* 32 sp 33 ! 34 " 35 # 36 $ 37 % 38 & 39 ' */ + 0, '!', 0, '#', '$', '%', '&', '\'', +/* 40 ( 41 ) 42 * 43 + 44 , 45 - 46 . 47 / */ + 0, 0, '*', '+', 0, '-', '.', 0, +/* 48 0 49 1 50 2 51 3 52 4 53 5 54 6 55 7 */ + '0', '1', '2', '3', '4', '5', '6', '7', +/* 56 8 57 9 58 : 59 ; 60 < 61 = 62 > 63 ? */ + '8', '9', 0, 0, 0, 0, 0, 0, +/* 64 @ 65 A 66 B 67 C 68 D 69 E 70 F 71 G */ + 0, 'a', 'b', 'c', 'd', 'e', 'f', 'g', +/* 72 H 73 I 74 J 75 K 76 L 77 M 78 N 79 O */ + 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', +/* 80 P 81 Q 82 R 83 S 84 T 85 U 86 V 87 W */ + 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', +/* 88 X 89 Y 90 Z 91 [ 92 \ 93 ] 94 ^ 95 _ */ + 'x', 'y', 'z', 0, 0, 0, '^', '_', +/* 96 ` 97 a 98 b 99 c 100 d 101 e 102 f 103 g */ + '`', 'a', 'b', 'c', 'd', 'e', 'f', 'g', +/* 104 h 105 i 106 j 107 k 108 l 109 m 110 n 111 o */ + 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', +/* 112 p 113 q 114 r 115 s 116 t 117 u 118 v 119 w */ + 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', +/* 120 x 121 y 122 z 123 { 124 | 125 } 126 ~ 127 del */ + 'x', 'y', 'z', 0, '|', 0, '~', 0 }; + + +static const int8_t unhex[256] = + {-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1 + ,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1 + ,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1 + , 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,-1,-1,-1,-1,-1,-1 + ,-1,10,11,12,13,14,15,-1,-1,-1,-1,-1,-1,-1,-1,-1 + ,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1 + ,-1,10,11,12,13,14,15,-1,-1,-1,-1,-1,-1,-1,-1,-1 + ,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1 + }; + + + + char c, ch; + int8_t unhex_val; + const char *p = data; + const char *header_field_mark = 0; + const char *header_value_mark = 0; + const char *url_mark = 0; + const char *body_mark = 0; + const char *status_mark = 0; + + /* We're in an error state. Don't bother doing anything. */ + if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { + return 0; + } + + if (len == 0) { + switch (parser->state) { + case s_body_identity_eof: + /* Use of CALLBACK_NOTIFY() here would erroneously return 1 byte read if + * we got paused. + */ + CALLBACK_NOTIFY_NOADVANCE(message_complete); + return 0; + + case s_dead: + case s_start_req_or_res: + case s_start_res: + case s_start_req: + return 0; + + default: + SET_ERRNO(HPE_INVALID_EOF_STATE); + return 1; + } + } + + + if (parser->state == s_header_field) + header_field_mark = data; + if (parser->state == s_header_value) + header_value_mark = data; + switch (parser->state) { + case s_req_path: + case s_req_schema: + case s_req_schema_slash: + case s_req_schema_slash_slash: + case s_req_server_start: + case s_req_server: + case s_req_server_with_at: + case s_req_query_string_start: + case s_req_query_string: + case s_req_fragment_start: + case s_req_fragment: + url_mark = data; + break; + case s_res_status: + status_mark = data; + break; + } + + for (p=data; p != data + len; p++) { + ch = *p; + + if (PARSING_HEADER(parser->state)) { + ++parser->nread; + /* Don't allow the total size of the HTTP headers (including the status + * line) to exceed HTTP_MAX_HEADER_SIZE. This check is here to protect + * embedders against denial-of-service attacks where the attacker feeds + * us a never-ending header that the embedder keeps buffering. + * + * This check is arguably the responsibility of embedders but we're doing + * it on the embedder's behalf because most won't bother and this way we + * make the web a little safer. HTTP_MAX_HEADER_SIZE is still far bigger + * than any reasonable request or response so this should never affect + * day-to-day operation. + */ + if (parser->nread > (HTTP_MAX_HEADER_SIZE)) { + SET_ERRNO(HPE_HEADER_OVERFLOW); + goto error; + } + } + + reexecute_byte: + switch (parser->state) { + + case s_dead: + /* this state is used after a 'Connection: close' message + * the parser will error out if it reads another message + */ + if (ch == CR || ch == LF) + break; + + SET_ERRNO(HPE_CLOSED_CONNECTION); + goto error; + + case s_start_req_or_res: + { + if (ch == CR || ch == LF) + break; + parser->flags = 0; + parser->content_length = ULLONG_MAX; + + if (ch == 'H') { + parser->state = s_res_or_resp_H; + + CALLBACK_NOTIFY(message_begin); + } else { + parser->type = HTTP_REQUEST; + parser->state = s_start_req; + goto reexecute_byte; + } + + break; + } + + case s_res_or_resp_H: + if (ch == 'T') { + parser->type = HTTP_RESPONSE; + parser->state = s_res_HT; + } else { + if (ch != 'E') { + SET_ERRNO(HPE_INVALID_CONSTANT); + goto error; + } + + parser->type = HTTP_REQUEST; + parser->method = HTTP_HEAD; + parser->index = 2; + parser->state = s_req_method; + } + break; + + case s_start_res: + { + parser->flags = 0; + parser->content_length = ULLONG_MAX; + + switch (ch) { + case 'H': + parser->state = s_res_H; + break; + + case CR: + case LF: + break; + + default: + SET_ERRNO(HPE_INVALID_CONSTANT); + goto error; + } + + CALLBACK_NOTIFY(message_begin); + break; + } + + case s_res_H: + STRICT_CHECK(ch != 'T'); + parser->state = s_res_HT; + break; + + case s_res_HT: + STRICT_CHECK(ch != 'T'); + parser->state = s_res_HTT; + break; + + case s_res_HTT: + STRICT_CHECK(ch != 'P'); + parser->state = s_res_HTTP; + break; + + case s_res_HTTP: + STRICT_CHECK(ch != '/'); + parser->state = s_res_first_http_major; + break; + + case s_res_first_http_major: + if (ch < '0' || ch > '9') { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_major = ch - '0'; + parser->state = s_res_http_major; + break; + + /* major HTTP version or dot */ + case s_res_http_major: + { + if (ch == '.') { + parser->state = s_res_first_http_minor; + break; + } + + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_major *= 10; + parser->http_major += ch - '0'; + + if (parser->http_major > 999) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + break; + } + + /* first digit of minor HTTP version */ + case s_res_first_http_minor: + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_minor = ch - '0'; + parser->state = s_res_http_minor; + break; + + /* minor HTTP version or end of request line */ + case s_res_http_minor: + { + if (ch == ' ') { + parser->state = s_res_first_status_code; + break; + } + + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_minor *= 10; + parser->http_minor += ch - '0'; + + if (parser->http_minor > 999) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + break; + } + + case s_res_first_status_code: + { + if (!IS_NUM(ch)) { + if (ch == ' ') { + break; + } + + SET_ERRNO(HPE_INVALID_STATUS); + goto error; + } + parser->status_code = ch - '0'; + parser->state = s_res_status_code; + break; + } + + case s_res_status_code: + { + if (!IS_NUM(ch)) { + switch (ch) { + case ' ': + parser->state = s_res_status_start; + break; + case CR: + parser->state = s_res_line_almost_done; + break; + case LF: + parser->state = s_header_field_start; + break; + default: + SET_ERRNO(HPE_INVALID_STATUS); + goto error; + } + break; + } + + parser->status_code *= 10; + parser->status_code += ch - '0'; + + if (parser->status_code > 999) { + SET_ERRNO(HPE_INVALID_STATUS); + goto error; + } + + break; + } + + case s_res_status_start: + { + if (ch == CR) { + parser->state = s_res_line_almost_done; + break; + } + + if (ch == LF) { + parser->state = s_header_field_start; + break; + } + + MARK(status); + parser->state = s_res_status; + parser->index = 0; + break; + } + + case s_res_status: + if (ch == CR) { + parser->state = s_res_line_almost_done; + CALLBACK_DATA(status); + break; + } + + if (ch == LF) { + parser->state = s_header_field_start; + CALLBACK_DATA(status); + break; + } + + break; + + case s_res_line_almost_done: + STRICT_CHECK(ch != LF); + parser->state = s_header_field_start; + break; + + case s_start_req: + { + if (ch == CR || ch == LF) + break; + parser->flags = 0; + parser->content_length = ULLONG_MAX; + + if (!IS_ALPHA(ch)) { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + + parser->method = (enum http_method) 0; + parser->index = 1; + switch (ch) { + case 'C': parser->method = HTTP_CONNECT; /* or COPY, CHECKOUT */ break; + case 'D': parser->method = HTTP_DELETE; break; + case 'G': parser->method = HTTP_GET; break; + case 'H': parser->method = HTTP_HEAD; break; + case 'L': parser->method = HTTP_LOCK; break; + case 'M': parser->method = HTTP_MKCOL; /* or MOVE, MKACTIVITY, MERGE, M-SEARCH, MKCALENDAR */ break; + case 'N': parser->method = HTTP_NOTIFY; break; + case 'O': parser->method = HTTP_OPTIONS; break; + case 'P': parser->method = HTTP_POST; + /* or PROPFIND|PROPPATCH|PUT|PATCH|PURGE */ + break; + case 'R': parser->method = HTTP_REPORT; break; + case 'S': parser->method = HTTP_SUBSCRIBE; /* or SEARCH */ break; + case 'T': parser->method = HTTP_TRACE; break; + case 'U': parser->method = HTTP_UNLOCK; /* or UNSUBSCRIBE */ break; + default: + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + parser->state = s_req_method; + + CALLBACK_NOTIFY(message_begin); + + break; + } + + case s_req_method: + { + const char *matcher; + if (ch == '\0') { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + + matcher = method_strings[parser->method]; + if (ch == ' ' && matcher[parser->index] == '\0') { + parser->state = s_req_spaces_before_url; + } else if (ch == matcher[parser->index]) { + ; /* nada */ + } else if (parser->method == HTTP_CONNECT) { + if (parser->index == 1 && ch == 'H') { + parser->method = HTTP_CHECKOUT; + } else if (parser->index == 2 && ch == 'P') { + parser->method = HTTP_COPY; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->method == HTTP_MKCOL) { + if (parser->index == 1 && ch == 'O') { + parser->method = HTTP_MOVE; + } else if (parser->index == 1 && ch == 'E') { + parser->method = HTTP_MERGE; + } else if (parser->index == 1 && ch == '-') { + parser->method = HTTP_MSEARCH; + } else if (parser->index == 2 && ch == 'A') { + parser->method = HTTP_MKACTIVITY; + } else if (parser->index == 3 && ch == 'A') { + parser->method = HTTP_MKCALENDAR; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->method == HTTP_SUBSCRIBE) { + if (parser->index == 1 && ch == 'E') { + parser->method = HTTP_SEARCH; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->index == 1 && parser->method == HTTP_POST) { + if (ch == 'R') { + parser->method = HTTP_PROPFIND; /* or HTTP_PROPPATCH */ + } else if (ch == 'U') { + parser->method = HTTP_PUT; /* or HTTP_PURGE */ + } else if (ch == 'A') { + parser->method = HTTP_PATCH; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->index == 2) { + if (parser->method == HTTP_PUT) { + if (ch == 'R') { + parser->method = HTTP_PURGE; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->method == HTTP_UNLOCK) { + if (ch == 'S') { + parser->method = HTTP_UNSUBSCRIBE; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + } else if (parser->index == 4 && parser->method == HTTP_PROPFIND && ch == 'P') { + parser->method = HTTP_PROPPATCH; + } else { + SET_ERRNO(HPE_INVALID_METHOD); + goto error; + } + + ++parser->index; + break; + } + + case s_req_spaces_before_url: + { + if (ch == ' ') break; + + MARK(url); + if (parser->method == HTTP_CONNECT) { + parser->state = s_req_server_start; + } + + parser->state = parse_url_char((enum state)parser->state, ch); + if (parser->state == s_dead) { + SET_ERRNO(HPE_INVALID_URL); + goto error; + } + + break; + } + + case s_req_schema: + case s_req_schema_slash: + case s_req_schema_slash_slash: + case s_req_server_start: + { + switch (ch) { + /* No whitespace allowed here */ + case ' ': + case CR: + case LF: + SET_ERRNO(HPE_INVALID_URL); + goto error; + default: + parser->state = parse_url_char((enum state)parser->state, ch); + if (parser->state == s_dead) { + SET_ERRNO(HPE_INVALID_URL); + goto error; + } + } + + break; + } + + case s_req_server: + case s_req_server_with_at: + case s_req_path: + case s_req_query_string_start: + case s_req_query_string: + case s_req_fragment_start: + case s_req_fragment: + { + switch (ch) { + case ' ': + parser->state = s_req_http_start; + CALLBACK_DATA(url); + break; + case CR: + case LF: + parser->http_major = 0; + parser->http_minor = 9; + parser->state = (ch == CR) ? + s_req_line_almost_done : + s_header_field_start; + CALLBACK_DATA(url); + break; + default: + parser->state = parse_url_char((enum state)parser->state, ch); + if (parser->state == s_dead) { + SET_ERRNO(HPE_INVALID_URL); + goto error; + } + } + break; + } + + case s_req_http_start: + switch (ch) { + case 'H': + parser->state = s_req_http_H; + break; + case ' ': + break; + default: + SET_ERRNO(HPE_INVALID_CONSTANT); + goto error; + } + break; + + case s_req_http_H: + STRICT_CHECK(ch != 'T'); + parser->state = s_req_http_HT; + break; + + case s_req_http_HT: + STRICT_CHECK(ch != 'T'); + parser->state = s_req_http_HTT; + break; + + case s_req_http_HTT: + STRICT_CHECK(ch != 'P'); + parser->state = s_req_http_HTTP; + break; + + case s_req_http_HTTP: + STRICT_CHECK(ch != '/'); + parser->state = s_req_first_http_major; + break; + + /* first digit of major HTTP version */ + case s_req_first_http_major: + if (ch < '1' || ch > '9') { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_major = ch - '0'; + parser->state = s_req_http_major; + break; + + /* major HTTP version or dot */ + case s_req_http_major: + { + if (ch == '.') { + parser->state = s_req_first_http_minor; + break; + } + + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_major *= 10; + parser->http_major += ch - '0'; + + if (parser->http_major > 999) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + break; + } + + /* first digit of minor HTTP version */ + case s_req_first_http_minor: + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_minor = ch - '0'; + parser->state = s_req_http_minor; + break; + + /* minor HTTP version or end of request line */ + case s_req_http_minor: + { + if (ch == CR) { + parser->state = s_req_line_almost_done; + break; + } + + if (ch == LF) { + parser->state = s_header_field_start; + break; + } + + /* XXX allow spaces after digit? */ + + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + parser->http_minor *= 10; + parser->http_minor += ch - '0'; + + if (parser->http_minor > 999) { + SET_ERRNO(HPE_INVALID_VERSION); + goto error; + } + + break; + } + + /* end of request line */ + case s_req_line_almost_done: + { + if (ch != LF) { + SET_ERRNO(HPE_LF_EXPECTED); + goto error; + } + + parser->state = s_header_field_start; + break; + } + + case s_header_field_start: + { + if (ch == CR) { + parser->state = s_headers_almost_done; + break; + } + + if (ch == LF) { + /* they might be just sending \n instead of \r\n so this would be + * the second \n to denote the end of headers*/ + parser->state = s_headers_almost_done; + goto reexecute_byte; + } + + c = TOKEN(ch); + + if (!c) { + SET_ERRNO(HPE_INVALID_HEADER_TOKEN); + goto error; + } + + MARK(header_field); + + parser->index = 0; + parser->state = s_header_field; + + switch (c) { + case 'c': + parser->header_state = h_C; + break; + + case 'p': + parser->header_state = h_matching_proxy_connection; + break; + + case 't': + parser->header_state = h_matching_transfer_encoding; + break; + + case 'u': + parser->header_state = h_matching_upgrade; + break; + + default: + parser->header_state = h_general; + break; + } + break; + } + + case s_header_field: + { + c = TOKEN(ch); + + if (c) { + switch (parser->header_state) { + case h_general: + break; + + case h_C: + parser->index++; + parser->header_state = (c == 'o' ? h_CO : h_general); + break; + + case h_CO: + parser->index++; + parser->header_state = (c == 'n' ? h_CON : h_general); + break; + + case h_CON: + parser->index++; + switch (c) { + case 'n': + parser->header_state = h_matching_connection; + break; + case 't': + parser->header_state = h_matching_content_length; + break; + default: + parser->header_state = h_general; + break; + } + break; + + /* connection */ + + case h_matching_connection: + parser->index++; + if (parser->index > sizeof(CONNECTION)-1 + || c != CONNECTION[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(CONNECTION)-2) { + parser->header_state = h_connection; + } + break; + + /* proxy-connection */ + + case h_matching_proxy_connection: + parser->index++; + if (parser->index > sizeof(PROXY_CONNECTION)-1 + || c != PROXY_CONNECTION[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(PROXY_CONNECTION)-2) { + parser->header_state = h_connection; + } + break; + + /* content-length */ + + case h_matching_content_length: + parser->index++; + if (parser->index > sizeof(CONTENT_LENGTH)-1 + || c != CONTENT_LENGTH[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(CONTENT_LENGTH)-2) { + parser->header_state = h_content_length; + } + break; + + /* transfer-encoding */ + + case h_matching_transfer_encoding: + parser->index++; + if (parser->index > sizeof(TRANSFER_ENCODING)-1 + || c != TRANSFER_ENCODING[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(TRANSFER_ENCODING)-2) { + parser->header_state = h_transfer_encoding; + } + break; + + /* upgrade */ + + case h_matching_upgrade: + parser->index++; + if (parser->index > sizeof(UPGRADE)-1 + || c != UPGRADE[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(UPGRADE)-2) { + parser->header_state = h_upgrade; + } + break; + + case h_connection: + case h_content_length: + case h_transfer_encoding: + case h_upgrade: + if (ch != ' ') parser->header_state = h_general; + break; + + default: + assert(0 && "Unknown header_state"); + break; + } + break; + } + + if (ch == ':') { + parser->state = s_header_value_discard_ws; + CALLBACK_DATA(header_field); + break; + } + + if (ch == CR) { + parser->state = s_header_almost_done; + CALLBACK_DATA(header_field); + break; + } + + if (ch == LF) { + parser->state = s_header_field_start; + CALLBACK_DATA(header_field); + break; + } + + SET_ERRNO(HPE_INVALID_HEADER_TOKEN); + goto error; + } + + case s_header_value_discard_ws: + if (ch == ' ' || ch == '\t') break; + + if (ch == CR) { + parser->state = s_header_value_discard_ws_almost_done; + break; + } + + if (ch == LF) { + parser->state = s_header_value_discard_lws; + break; + } + + /* FALLTHROUGH */ + + case s_header_value_start: + { + MARK(header_value); + + parser->state = s_header_value; + parser->index = 0; + + c = LOWER(ch); + + switch (parser->header_state) { + case h_upgrade: + parser->flags |= F_UPGRADE; + parser->header_state = h_general; + break; + + case h_transfer_encoding: + /* looking for 'Transfer-Encoding: chunked' */ + if ('c' == c) { + parser->header_state = h_matching_transfer_encoding_chunked; + } else { + parser->header_state = h_general; + } + break; + + case h_content_length: + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_CONTENT_LENGTH); + goto error; + } + + parser->content_length = ch - '0'; + break; + + case h_connection: + /* looking for 'Connection: keep-alive' */ + if (c == 'k') { + parser->header_state = h_matching_connection_keep_alive; + /* looking for 'Connection: close' */ + } else if (c == 'c') { + parser->header_state = h_matching_connection_close; + } else { + parser->header_state = h_general; + } + break; + + default: + parser->header_state = h_general; + break; + } + break; + } + + case s_header_value: + { + + if (ch == CR) { + parser->state = s_header_almost_done; + CALLBACK_DATA(header_value); + break; + } + + if (ch == LF) { + parser->state = s_header_almost_done; + CALLBACK_DATA_NOADVANCE(header_value); + goto reexecute_byte; + } + + c = LOWER(ch); + + switch (parser->header_state) { + case h_general: + break; + + case h_connection: + case h_transfer_encoding: + assert(0 && "Shouldn't get here."); + break; + + case h_content_length: + { + uint64_t t; + + if (ch == ' ') break; + + if (!IS_NUM(ch)) { + SET_ERRNO(HPE_INVALID_CONTENT_LENGTH); + goto error; + } + + t = parser->content_length; + t *= 10; + t += ch - '0'; + + /* Overflow? Test against a conservative limit for simplicity. */ + if ((ULLONG_MAX - 10) / 10 < parser->content_length) { + SET_ERRNO(HPE_INVALID_CONTENT_LENGTH); + goto error; + } + + parser->content_length = t; + break; + } + + /* Transfer-Encoding: chunked */ + case h_matching_transfer_encoding_chunked: + parser->index++; + if (parser->index > sizeof(CHUNKED)-1 + || c != CHUNKED[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(CHUNKED)-2) { + parser->header_state = h_transfer_encoding_chunked; + } + break; + + /* looking for 'Connection: keep-alive' */ + case h_matching_connection_keep_alive: + parser->index++; + if (parser->index > sizeof(KEEP_ALIVE)-1 + || c != KEEP_ALIVE[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(KEEP_ALIVE)-2) { + parser->header_state = h_connection_keep_alive; + } + break; + + /* looking for 'Connection: close' */ + case h_matching_connection_close: + parser->index++; + if (parser->index > sizeof(CLOSE)-1 || c != CLOSE[parser->index]) { + parser->header_state = h_general; + } else if (parser->index == sizeof(CLOSE)-2) { + parser->header_state = h_connection_close; + } + break; + + case h_transfer_encoding_chunked: + case h_connection_keep_alive: + case h_connection_close: + if (ch != ' ') parser->header_state = h_general; + break; + + default: + parser->state = s_header_value; + parser->header_state = h_general; + break; + } + break; + } + + case s_header_almost_done: + { + STRICT_CHECK(ch != LF); + + parser->state = s_header_value_lws; + break; + } + + case s_header_value_lws: + { + if (ch == ' ' || ch == '\t') { + parser->state = s_header_value_start; + goto reexecute_byte; + } + + /* finished the header */ + switch (parser->header_state) { + case h_connection_keep_alive: + parser->flags |= F_CONNECTION_KEEP_ALIVE; + break; + case h_connection_close: + parser->flags |= F_CONNECTION_CLOSE; + break; + case h_transfer_encoding_chunked: + parser->flags |= F_CHUNKED; + break; + default: + break; + } + + parser->state = s_header_field_start; + goto reexecute_byte; + } + + case s_header_value_discard_ws_almost_done: + { + STRICT_CHECK(ch != LF); + parser->state = s_header_value_discard_lws; + break; + } + + case s_header_value_discard_lws: + { + if (ch == ' ' || ch == '\t') { + parser->state = s_header_value_discard_ws; + break; + } else { + /* header value was empty */ + MARK(header_value); + parser->state = s_header_field_start; + CALLBACK_DATA_NOADVANCE(header_value); + goto reexecute_byte; + } + } + + case s_headers_almost_done: + { + STRICT_CHECK(ch != LF); + + if (parser->flags & F_TRAILING) { + /* End of a chunked request */ + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + break; + } + + parser->state = s_headers_done; + + /* Set this here so that on_headers_complete() callbacks can see it */ + parser->upgrade = + (parser->flags & F_UPGRADE || parser->method == HTTP_CONNECT); + + /* Here we call the headers_complete callback. This is somewhat + * different than other callbacks because if the user returns 1, we + * will interpret that as saying that this message has no body. This + * is needed for the annoying case of recieving a response to a HEAD + * request. + * + * We'd like to use CALLBACK_NOTIFY_NOADVANCE() here but we cannot, so + * we have to simulate it by handling a change in errno below. + */ + if (settings->on_headers_complete) { + switch (settings->on_headers_complete(parser)) { + case 0: + break; + + case 1: + parser->flags |= F_SKIPBODY; + break; + + default: + SET_ERRNO(HPE_CB_headers_complete); + return p - data; /* Error */ + } + } + + if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { + return p - data; + } + + goto reexecute_byte; + } + + case s_headers_done: + { + STRICT_CHECK(ch != LF); + + parser->nread = 0; + + /* Exit, the rest of the connect is in a different protocol. */ + if (parser->upgrade) { + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + return (p - data) + 1; + } + + if (parser->flags & F_SKIPBODY) { + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + } else if (parser->flags & F_CHUNKED) { + /* chunked encoding - ignore Content-Length header */ + parser->state = s_chunk_size_start; + } else { + if (parser->content_length == 0) { + /* Content-Length header given but zero: Content-Length: 0\r\n */ + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + } else if (parser->content_length != ULLONG_MAX) { + /* Content-Length header given and non-zero */ + parser->state = s_body_identity; + } else { + if (parser->type == HTTP_REQUEST || + !http_message_needs_eof(parser)) { + /* Assume content-length 0 - read the next */ + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + } else { + /* Read body until EOF */ + parser->state = s_body_identity_eof; + } + } + } + + break; + } + + case s_body_identity: + { + uint64_t to_read = MIN(parser->content_length, + (uint64_t) ((data + len) - p)); + + assert(parser->content_length != 0 + && parser->content_length != ULLONG_MAX); + + /* The difference between advancing content_length and p is because + * the latter will automaticaly advance on the next loop iteration. + * Further, if content_length ends up at 0, we want to see the last + * byte again for our message complete callback. + */ + MARK(body); + parser->content_length -= to_read; + p += to_read - 1; + + if (parser->content_length == 0) { + parser->state = s_message_done; + + /* Mimic CALLBACK_DATA_NOADVANCE() but with one extra byte. + * + * The alternative to doing this is to wait for the next byte to + * trigger the data callback, just as in every other case. The + * problem with this is that this makes it difficult for the test + * harness to distinguish between complete-on-EOF and + * complete-on-length. It's not clear that this distinction is + * important for applications, but let's keep it for now. + */ + CALLBACK_DATA_(body, p - body_mark + 1, p - data); + goto reexecute_byte; + } + + break; + } + + /* read until EOF */ + case s_body_identity_eof: + MARK(body); + p = data + len - 1; + + break; + + case s_message_done: + parser->state = NEW_MESSAGE(); + CALLBACK_NOTIFY(message_complete); + break; + + case s_chunk_size_start: + { + assert(parser->nread == 1); + assert(parser->flags & F_CHUNKED); + + unhex_val = unhex[(unsigned char)ch]; + if (unhex_val == -1) { + SET_ERRNO(HPE_INVALID_CHUNK_SIZE); + goto error; + } + + parser->content_length = unhex_val; + parser->state = s_chunk_size; + break; + } + + case s_chunk_size: + { + uint64_t t; + + assert(parser->flags & F_CHUNKED); + + if (ch == CR) { + parser->state = s_chunk_size_almost_done; + break; + } + + unhex_val = unhex[(unsigned char)ch]; + + if (unhex_val == -1) { + if (ch == ';' || ch == ' ') { + parser->state = s_chunk_parameters; + break; + } + + SET_ERRNO(HPE_INVALID_CHUNK_SIZE); + goto error; + } + + t = parser->content_length; + t *= 16; + t += unhex_val; + + /* Overflow? Test against a conservative limit for simplicity. */ + if ((ULLONG_MAX - 16) / 16 < parser->content_length) { + SET_ERRNO(HPE_INVALID_CONTENT_LENGTH); + goto error; + } + + parser->content_length = t; + break; + } + + case s_chunk_parameters: + { + assert(parser->flags & F_CHUNKED); + /* just ignore this shit. TODO check for overflow */ + if (ch == CR) { + parser->state = s_chunk_size_almost_done; + break; + } + break; + } + + case s_chunk_size_almost_done: + { + assert(parser->flags & F_CHUNKED); + STRICT_CHECK(ch != LF); + + parser->nread = 0; + + if (parser->content_length == 0) { + parser->flags |= F_TRAILING; + parser->state = s_header_field_start; + } else { + parser->state = s_chunk_data; + } + break; + } + + case s_chunk_data: + { + uint64_t to_read = MIN(parser->content_length, + (uint64_t) ((data + len) - p)); + + assert(parser->flags & F_CHUNKED); + assert(parser->content_length != 0 + && parser->content_length != ULLONG_MAX); + + /* See the explanation in s_body_identity for why the content + * length and data pointers are managed this way. + */ + MARK(body); + parser->content_length -= to_read; + p += to_read - 1; + + if (parser->content_length == 0) { + parser->state = s_chunk_data_almost_done; + } + + break; + } + + case s_chunk_data_almost_done: + assert(parser->flags & F_CHUNKED); + assert(parser->content_length == 0); + STRICT_CHECK(ch != CR); + parser->state = s_chunk_data_done; + CALLBACK_DATA(body); + break; + + case s_chunk_data_done: + assert(parser->flags & F_CHUNKED); + STRICT_CHECK(ch != LF); + parser->nread = 0; + parser->state = s_chunk_size_start; + break; + + default: + assert(0 && "unhandled state"); + SET_ERRNO(HPE_INVALID_INTERNAL_STATE); + goto error; + } + } + + /* Run callbacks for any marks that we have leftover after we ran our of + * bytes. There should be at most one of these set, so it's OK to invoke + * them in series (unset marks will not result in callbacks). + * + * We use the NOADVANCE() variety of callbacks here because 'p' has already + * overflowed 'data' and this allows us to correct for the off-by-one that + * we'd otherwise have (since CALLBACK_DATA() is meant to be run with a 'p' + * value that's in-bounds). + */ + + assert(((header_field_mark ? 1 : 0) + + (header_value_mark ? 1 : 0) + + (url_mark ? 1 : 0) + + (body_mark ? 1 : 0) + + (status_mark ? 1 : 0)) <= 1); + + CALLBACK_DATA_NOADVANCE(header_field); + CALLBACK_DATA_NOADVANCE(header_value); + CALLBACK_DATA_NOADVANCE(url); + CALLBACK_DATA_NOADVANCE(body); + CALLBACK_DATA_NOADVANCE(status); + + return len; + +error: + if (HTTP_PARSER_ERRNO(parser) == HPE_OK) { + SET_ERRNO(HPE_UNKNOWN); + } + + return (p - data); +} + + +/* Does the parser need to see an EOF to find the end of the message? */ +int +http_message_needs_eof (const http_parser *parser) +{ + if (parser->type == HTTP_REQUEST) { + return 0; + } + + /* See RFC 2616 section 4.4 */ + if (parser->status_code / 100 == 1 || /* 1xx e.g. Continue */ + parser->status_code == 204 || /* No Content */ + parser->status_code == 304 || /* Not Modified */ + parser->flags & F_SKIPBODY) { /* response to a HEAD request */ + return 0; + } + + if ((parser->flags & F_CHUNKED) || parser->content_length != ULLONG_MAX) { + return 0; + } + + return 1; +} + + +int +http_should_keep_alive (const http_parser *parser) +{ + if (parser->http_major > 0 && parser->http_minor > 0) { + /* HTTP/1.1 */ + if (parser->flags & F_CONNECTION_CLOSE) { + return 0; + } + } else { + /* HTTP/1.0 or earlier */ + if (!(parser->flags & F_CONNECTION_KEEP_ALIVE)) { + return 0; + } + } + + return !http_message_needs_eof(parser); +} + + +const char * +http_method_str (enum http_method m) +{ +static const char *method_strings[] = + { +#define XX(num, name, string) #string, + HTTP_METHOD_MAP(XX) +#undef XX + }; + return ELEM_AT(method_strings, m, ""); +} + + +void +http_parser_init (http_parser *parser, enum http_parser_type t) +{ + void *data = parser->data; /* preserve application data */ + memset(parser, 0, sizeof(*parser)); + parser->data = data; + parser->type = t; + parser->state = (t == HTTP_REQUEST ? s_start_req : (t == HTTP_RESPONSE ? s_start_res : s_start_req_or_res)); + parser->http_errno = HPE_OK; +} + +const char * +http_errno_name(enum http_errno err) { +/* Map errno values to strings for human-readable output */ +#define HTTP_STRERROR_GEN(n, s) { "HPE_" #n, s }, +static struct { + const char *name; + const char *description; +} http_strerror_tab[] = { + HTTP_ERRNO_MAP(HTTP_STRERROR_GEN) +}; +#undef HTTP_STRERROR_GEN + assert(err < (sizeof(http_strerror_tab)/sizeof(http_strerror_tab[0]))); + return http_strerror_tab[err].name; +} + +const char * +http_errno_description(enum http_errno err) { +/* Map errno values to strings for human-readable output */ +#define HTTP_STRERROR_GEN(n, s) { "HPE_" #n, s }, +static struct { + const char *name; + const char *description; +} http_strerror_tab[] = { + HTTP_ERRNO_MAP(HTTP_STRERROR_GEN) +}; +#undef HTTP_STRERROR_GEN + assert(err < (sizeof(http_strerror_tab)/sizeof(http_strerror_tab[0]))); + return http_strerror_tab[err].description; +} + +static enum http_host_state +http_parse_host_char(enum http_host_state s, const char ch) { + switch(s) { + case s_http_userinfo: + case s_http_userinfo_start: + if (ch == '@') { + return s_http_host_start; + } + + if (IS_USERINFO_CHAR(ch)) { + return s_http_userinfo; + } + break; + + case s_http_host_start: + if (ch == '[') { + return s_http_host_v6_start; + } + + if (IS_HOST_CHAR(ch)) { + return s_http_host; + } + + break; + + case s_http_host: + if (IS_HOST_CHAR(ch)) { + return s_http_host; + } + + /* FALLTHROUGH */ + case s_http_host_v6_end: + if (ch == ':') { + return s_http_host_port_start; + } + + break; + + case s_http_host_v6: + if (ch == ']') { + return s_http_host_v6_end; + } + + /* FALLTHROUGH */ + case s_http_host_v6_start: + if (IS_HEX(ch) || ch == ':' || ch == '.') { + return s_http_host_v6; + } + + break; + + case s_http_host_port: + case s_http_host_port_start: + if (IS_NUM(ch)) { + return s_http_host_port; + } + + break; + + default: + break; + } + return s_http_host_dead; +} + +static int +http_parse_host(const char * buf, struct http_parser_url *u, int found_at) { + enum http_host_state s; + + const char *p; + size_t buflen = u->field_data[UF_HOST].off + u->field_data[UF_HOST].len; + + u->field_data[UF_HOST].len = 0; + + s = found_at ? s_http_userinfo_start : s_http_host_start; + + for (p = buf + u->field_data[UF_HOST].off; p < buf + buflen; p++) { + enum http_host_state new_s = http_parse_host_char(s, *p); + + if (new_s == s_http_host_dead) { + return 1; + } + + switch(new_s) { + case s_http_host: + if (s != s_http_host) { + u->field_data[UF_HOST].off = p - buf; + } + u->field_data[UF_HOST].len++; + break; + + case s_http_host_v6: + if (s != s_http_host_v6) { + u->field_data[UF_HOST].off = p - buf; + } + u->field_data[UF_HOST].len++; + break; + + case s_http_host_port: + if (s != s_http_host_port) { + u->field_data[UF_PORT].off = p - buf; + u->field_data[UF_PORT].len = 0; + u->field_set |= (1 << UF_PORT); + } + u->field_data[UF_PORT].len++; + break; + + case s_http_userinfo: + if (s != s_http_userinfo) { + u->field_data[UF_USERINFO].off = p - buf ; + u->field_data[UF_USERINFO].len = 0; + u->field_set |= (1 << UF_USERINFO); + } + u->field_data[UF_USERINFO].len++; + break; + + default: + break; + } + s = new_s; + } + + /* Make sure we don't end somewhere unexpected */ + switch (s) { + case s_http_host_start: + case s_http_host_v6_start: + case s_http_host_v6: + case s_http_host_port_start: + case s_http_userinfo: + case s_http_userinfo_start: + return 1; + default: + break; + } + + return 0; +} + +int +http_parser_parse_url(const char *buf, size_t buflen, int is_connect, + struct http_parser_url *u) +{ + enum state s; + const char *p; + enum http_parser_url_fields uf, old_uf; + int found_at = 0; + + u->port = u->field_set = 0; + s = is_connect ? s_req_server_start : s_req_spaces_before_url; + old_uf = UF_MAX; + + for (p = buf; p < buf + buflen; p++) { + s = parse_url_char(s, *p); + + /* Figure out the next field that we're operating on */ + switch (s) { + case s_dead: + return 1; + + /* Skip delimeters */ + case s_req_schema_slash: + case s_req_schema_slash_slash: + case s_req_server_start: + case s_req_query_string_start: + case s_req_fragment_start: + continue; + + case s_req_schema: + uf = UF_SCHEMA; + break; + + case s_req_server_with_at: + found_at = 1; + + /* FALLTROUGH */ + case s_req_server: + uf = UF_HOST; + break; + + case s_req_path: + uf = UF_PATH; + break; + + case s_req_query_string: + uf = UF_QUERY; + break; + + case s_req_fragment: + uf = UF_FRAGMENT; + break; + + default: + assert(!"Unexpected state"); + return 1; + } + + /* Nothing's changed; soldier on */ + if (uf == old_uf) { + u->field_data[uf].len++; + continue; + } + + u->field_data[uf].off = p - buf; + u->field_data[uf].len = 1; + + u->field_set |= (1 << uf); + old_uf = uf; + } + + /* host must be present if there is a schema */ + /* parsing http:///toto will fail */ + if ((u->field_set & ((1 << UF_SCHEMA) | (1 << UF_HOST))) != 0) { + if (http_parse_host(buf, u, found_at) != 0) { + return 1; + } + } + + /* CONNECT requests can only contain "hostname:port" */ + if (is_connect && u->field_set != ((1 << UF_HOST)|(1 << UF_PORT))) { + return 1; + } + + if (u->field_set & (1 << UF_PORT)) { + /* Don't bother with endp; we've already validated the string */ + unsigned long v = strtoul(buf + u->field_data[UF_PORT].off, NULL, 10); + + /* Ports have a max value of 2^16 */ + if (v > 0xffff) { + return 1; + } + + u->port = (uint16_t) v; + } + + return 0; +} + +void +http_parser_pause(http_parser *parser, int paused) { + /* Users should only be pausing/unpausing a parser that is not in an error + * state. In non-debug builds, there's not much that we can do about this + * other than ignore it. + */ + if (HTTP_PARSER_ERRNO(parser) == HPE_OK || + HTTP_PARSER_ERRNO(parser) == HPE_PAUSED) { + SET_ERRNO((paused) ? HPE_PAUSED : HPE_OK); + } else { + assert(0 && "Attempting to pause parser in error state"); + } +} + +int +http_body_is_final(const struct http_parser *parser) { + return parser->state == s_message_done; +} + +unsigned long +http_parser_version(void) { + return HTTP_PARSER_VERSION_MAJOR * 0x10000 | + HTTP_PARSER_VERSION_MINOR * 0x00100 | + HTTP_PARSER_VERSION_PATCH * 0x00001; +} + +#undef HTTP_METHOD_MAP +#undef HTTP_ERRNO_MAP +#undef SET_ERRNO +#undef CALLBACK_NOTIFY_ +#undef CALLBACK_NOTIFY +#undef CALLBACK_NOTIFY_NOADVANCE +#undef CALLBACK_DATA_ +#undef CALLBACK_DATA +#undef CALLBACK_DATA_NOADVANCE +#undef MARK +#undef PROXY_CONNECTION +#undef CONNECTION +#undef CONTENT_LENGTH +#undef TRANSFER_ENCODING +#undef UPGRADE +#undef CHUNKED +#undef KEEP_ALIVE +#undef CLOSE +#undef PARSING_HEADER +#undef CR +#undef LF +#undef LOWER +#undef IS_ALPHA +#undef IS_NUM +#undef IS_ALPHANUM +#undef IS_HEX +#undef IS_MARK +#undef IS_USERINFO_CHAR +#undef TOKEN +#undef IS_URL_CHAR +#undef IS_HOST_CHAR +#undef start_state +#undef STRICT_CHECK +#undef NEW_MESSAGE + +#ifdef __cplusplus +} +#endif +#endif diff --git a/include/mustache.h b/include/mustache.h index fb4d664..6fd210b 100644 --- a/include/mustache.h +++ b/include/mustache.h @@ -528,7 +528,7 @@ namespace crow namespace detail { - inline std::function get_loader_ref() + inline std::function& get_loader_ref() { static std::function loader = default_loader; return loader; diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt deleted file mode 100644 index 4ac6f63..0000000 --- a/src/CMakeLists.txt +++ /dev/null @@ -1,17 +0,0 @@ -cmake_minimum_required(VERSION 2.8) -project (crow) - -set(CROW_SRCS -#${PROJECT_SOURCE_DIR}/some.cpp -#${PROJECT_SOURCE_DIR}/someother.cpp -${PROJECT_SOURCE_DIR}/../http-parser/http_parser.c -) - -set_source_files_properties(${PROJECT_SOURCE_DIR}/../http-parser/http_parser.c PROPERTIES LANGUAGE C ) - -include_directories("${PROJECT_BINARY_DIR}") -include_directories("${PROJECT_INCLUDE_DIR}") - - -add_library(${PROJECT_NAME} SHARED ${CROW_SRCS}) -#target_link_libraries(${PROJECT_NAME} tcmalloc) diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt index d35925c..6beb0be 100644 --- a/tests/CMakeLists.txt +++ b/tests/CMakeLists.txt @@ -7,7 +7,7 @@ unittest.cpp ) add_executable(unittest ${TEST_SRCS}) -target_link_libraries(unittest crow) +#target_link_libraries(unittest crow) target_link_libraries(unittest ${Boost_LIBRARIES} ) set_target_properties(unittest PROPERTIES COMPILE_FLAGS "--coverage -fprofile-arcs -ftest-coverage") diff --git a/tests/template/CMakeLists.txt b/tests/template/CMakeLists.txt index 1d4ec30..86c5d25 100644 --- a/tests/template/CMakeLists.txt +++ b/tests/template/CMakeLists.txt @@ -15,8 +15,8 @@ add_executable(mustachetest ${TEST_SRCS}) #target_link_libraries(unittest ${Boost_LIBRARIES} ) set_target_properties(mustachetest PROPERTIES COMPILE_FLAGS "-Wall -std=c++1y") -message(${PROJECT_SOURCE_DIR}) -message(${CMAKE_CURRENT_BINARY_DIR}) +#message(${PROJECT_SOURCE_DIR}) +#message(${CMAKE_CURRENT_BINARY_DIR}) file(COPY DIRECTORY . DESTINATION ${CMAKE_CURRENT_BINARY_DIR} FILES_MATCHING PATTERN "*.json") -- cgit v1.2.3-54-g00ecf