Merge pull request #3527 from elemoine/makefile

Replace pake with make?
This commit is contained in:
Éric Lemoine
2015-04-26 21:12:09 +02:00
17 changed files with 753 additions and 1389 deletions

View File

@@ -9,8 +9,8 @@ before_script:
- "rm src/ol/renderer/webgl/*shader.js"
- "sh -e /etc/init.d/xvfb start"
script: "./build.py ci"
script: "make ci"
after_success:
- "npm run test-coverage"
- "make test-coverage"
- "cat coverage/lcov.info | ./node_modules/.bin/coveralls"

View File

@@ -58,16 +58,17 @@ as described below.
The minimum requirements are:
* GNU Make
* Git
* [Node.js](http://nodejs.org/) (0.10.x or higher)
* Python 2.6 or 2.7 with a couple of extra modules (see below)
* Java 7 (JRE and JDK)
The executables `git`, `java`, `jar`, and `python` should be in your `PATH`.
The executables `git`, `node`, `python` and `java` should be in your `PATH`.
You can check your configuration by running:
$ ./build.py checkdeps
$ make check-deps
To install the Node.js dependencies run
@@ -82,24 +83,22 @@ or
depending on your OS and Python installation.
(You can also install the Python modules in a Python virtual environment if you want to.)
## Working with the build tool
As an ol3 developer you will need to use the `build.py` Python script. This is
the script to use to run the linter, the compiler, the tests, etc. Windows users
can use `build.cmd` which is a thin wrapper around `build.py`.
As an ol3 developer you will use `make` to run build targets defined in the
`Makefile` located at the root of the repository. The `Makefile` includes
targets for running the linter, the compiler, the tests, etc.
The `build.py` script is equivalent to a Makefile. It is actually based on
[pake](https://github.com/twpayne/pake/), which is a simple implementation of
`make` in Python.
The usage of `make` is as follows:
The usage of the script is:
$ ./build.py <target>
$ make <target>
where `<target>` is the name of the build target you want to execute. For
example:
$ ./build.py test
$ make test
The main build targets are `serve`, `lint`, `build`, `test`, and `check`. The
latter is a meta-target that basically runs `lint`, `build`, and `test`.
@@ -116,7 +115,7 @@ and have therefore no chance of being merged into `master`.
To run the `check` target:
$ ./build.py check
$ make check
If you want to run the full suite of integration tests, see "Running the integration
tests" below.
@@ -125,29 +124,29 @@ tests" below.
To run the examples you first need to start the dev server:
$ ./build.py serve
$ make serve
Then, just point your browser <http://localhost:3000/examples> in your browser. For example <http://localhost:3000/examples/side-by-side.html>.
Then, just point your browser <http://localhost:3000/build/examples> in your browser. For example <http://localhost:3000/build/examples/side-by-side.html>.
Run examples against the `ol.js` standalone build:
The examples can also be run against the `ol.js` standalone lib, just like the examples
[hosted](http://openlayers.github.com/ol3/master/examples/) on GitHub. Start by
executing the `host-examples` build target:
The examples can also be run against the `ol.js` standalone build, just like
the examples [hosted](http://openlayers.org/en/master/examples/) on GitHub.
Start by executing the `host-examples` build target:
$ ./build.py host-examples
$ make host-examples
After running `host-examples` you can now open the examples index page in the browser, for example: <http://localhost/~elemoine/ol3/build/hosted/master/examples/>. (This assumes that the `hosted` directory is a web directory, served by Apache for example.)
After running `host-examples` you can now open the examples index page in the browser: <http://localhost:3000/build/hosted/master/examples/>. (This assumes that you still have the dev server running.)
Append `?mode=raw` to make the example work in full debug mode. In raw mode the OpenLayers and Closure Library scripts are loaded individually by the Closure Library's `base.js` script (which the example page loads and executes before any other script).
## Running tests
To run the tests in a browser start the dev server (`./build.py serve`) and open <http://localhost:3000/test/index.html> in the browser.
To run the tests in a browser start the dev server (`make serve`) and open <http://localhost:3000/test/index.html> in the browser.
To run the tests on the console (headless testing with PhantomJS) use the `test` target:
$ ./build.py test
$ make test
See also the test-specific [README](../master/test/README.md).
@@ -161,7 +160,7 @@ displayed in the pull request.
To run the full suite of integration tests use the `ci` target:
$ ./build.py ci
$ make ci
Running the full suite of integration tests currently takes 5-10 minutes.
@@ -174,13 +173,8 @@ Adding functionality often implies adding one or several examples. This
section provides explanations related to adding examples.
The examples are located in the `examples` directory. Adding a new example
implies creating two files in this directory, an `.html` file and a `.js` file.
See `examples/simple.html` and `examples/simple.js` for instance.
The `.html` file needs to include a script tag with
`loader.js?id=<example_name>` as its `src`. For example, if the two files for
the example are `myexample.js` and `myexample.html` then the script tag's `src`
should be set to `myexample`.
implies creating two or three files in this directory, an `.html` file, a `.js`
file, and, optionally, a `.css` file.
You can use `simple.js` and `simple.html` as templates for new examples.
@@ -218,7 +212,7 @@ Your pull request must:
It is strongly recommended that you run
$ ./build.py check
$ make check
before every commit. This will catch many problems quickly, and it is much
faster than waiting for the Travis CI integration tests to run.
@@ -238,9 +232,9 @@ Guide](http://google-styleguide.googlecode.com/svn/trunk/javascriptguide.xml).
This is checked using the [Closure
Linter](https://developers.google.com/closure/utilities/) in strict mode. You
can run the linter locally on your machine before committing using the `lint`
target to `build.py`:
target:
$ ./build.py lint
$ make lint
In addition to fixing problems identified by the linter, please also follow the
style of the existing OpenLayers 3 code, which includes:
@@ -279,7 +273,7 @@ The integration tests contain a number of automated checks to ensure that the
code follows the OpenLayers 3 style and does not break tests or examples. You
can run the integration tests locally using the `ci` target:
$ ./build.py ci
$ make ci
### Address a single issue or add a single item of functionality

321
Makefile Normal file
View File

@@ -0,0 +1,321 @@
OS := $(shell uname)
BRANCH := $(shell git rev-parse --abbrev-ref HEAD)
SRC_GLSL := $(shell find src -type f -name '*.glsl')
SRC_SHADER_JS := $(patsubst %.glsl,%shader.js,$(SRC_GLSL))
SRC_JS := $(filter-out $(SRC_SHADER_JS),$(shell find src -name '*.js'))
SRC_JSDOC = $(shell find src -type f -name '*.jsdoc')
SPEC_JS := $(shell find test/spec -type f -name '*.js')
SPEC_RENDERING_JS := $(shell find test_rendering/spec -name '*.js')
EXAMPLES := $(shell find examples -type f)
EXAMPLES_HTML := $(filter-out examples/index.html,$(shell find examples -maxdepth 1 -type f -name '*.html'))
EXAMPLES_JS := $(patsubst %.html,%.js,$(EXAMPLES_HTML))
BUILD_EXAMPLES := $(subst examples,build/examples,$(EXAMPLES))
BUILD_HOSTED := build/hosted/$(BRANCH)
BUILD_HOSTED_EXAMPLES := $(addprefix $(BUILD_HOSTED)/,$(EXAMPLES))
BUILD_HOSTED_EXAMPLES_JS := $(addprefix $(BUILD_HOSTED)/,$(EXAMPLES_JS))
CHECK_EXAMPLE_TIMESTAMPS = $(patsubst examples/%.html,build/timestamps/check-%-timestamp,$(EXAMPLES_HTML))
TASKS_JS := $(shell find tasks -name '*.js')
CLOSURE_LIB = $(shell node -e 'process.stdout.write(require("closure-util").getLibraryPath())')
ifeq ($(OS),Darwin)
STAT_COMPRESSED = stat -f ' compressed: %z bytes'
STAT_UNCOMPRESSED = stat -f 'uncompressed: %z bytes'
else
STAT_COMPRESSED = stat -c ' compressed: %s bytes'
STAT_UNCOMPRESSED = stat -c 'uncompressed: %s bytes'
endif
.PHONY: default
default: help
.PHONY: help
help:
@echo
@echo "The most common targets are:"
@echo
@echo "- install Install node dependencies"
@echo "- serve Start dev server for running examples and tests"
@echo "- test Run unit tests in the console"
@echo "- check Perform a number of checks on the code"
@echo "- clean Remove generated files"
@echo "- help Display this help message"
@echo
@echo "Other less frequently used targets are:"
@echo
@echo "- build Build ol.js, ol-debug.js, ol.js.map and ol.css"
@echo "- lint Check the code with the linter"
@echo "- ci Run the full continuous integration process"
@echo "- apidoc Build the API documentation using JSDoc"
@echo "- cleanall Remove all the build artefacts"
@echo "- check-deps Check if the required dependencies are installed"
@echo
.PHONY: apidoc
apidoc: build/timestamps/jsdoc-$(BRANCH)-timestamp
.PHONY: build
build: build/ol.css build/ol.js build/ol-debug.js build/ol.js.map
.PHONY: check
check: lint build/ol.js test
.PHONY: check-examples
check-examples: $(CHECK_EXAMPLE_TIMESTAMPS)
.PHONY: check-deps
check-deps: EXECUTABLES = git node python java
check-deps:
@for exe in $(EXECUTABLES) ;\
do \
which $${exe} > /dev/null && \
echo "Program $${exe} OK" || \
echo "Program $${exe} MISSING!" ;\
done ;\
.PHONY: ci
ci: lint build test test-rendering compile-examples check-examples apidoc
.PHONY: compile-examples
compile-examples: build/compiled-examples/all.combined.js
.PHONY: clean
clean:
rm -f build/timestamps/gjslint-timestamp
rm -f build/timestamps/jshint-timestamp
rm -f build/timestamps/check-*-timestamp
rm -f build/ol.css
rm -f build/ol.js
rm -f build/ol.js.map
rm -f build/ol-debug.js
rm -f build/test_requires.js
rm -f build/test_rendering_requires.js
rm -rf build/examples
rm -rf build/compiled-examples
rm -rf $(BUILD_HOSTED)
.PHONY: cleanall
cleanall:
rm -rf build
.PHONY: css
css: build/ol.css
.PHONY: examples
examples: $(BUILD_EXAMPLES)
.PHONY: install
install: build/timestamps/node-modules-timestamp
.PHONY: lint
lint: build/timestamps/gjslint-timestamp build/timestamps/jshint-timestamp \
build/timestamps/check-requires-timestamp \
build/timestamps/check-whitespace-timestamp
.PHONY: npm-install
npm-install: build/timestamps/node-modules-timestamp
.PHONY: shaders
shaders: $(SRC_SHADER_JS)
.PHONY: serve
serve: build/test_requires.js build/test_rendering_requires.js
node tasks/serve.js
.PHONY: test
test: build/timestamps/node-modules-timestamp build/test_requires.js
node tasks/test.js
.PHONY: test-coverage
test-coverage: build/timestamps/node-modules-timestamp
node tasks/test-coverage.js
.PHONY: test-rendering
test-rendering: build/timestamps/node-modules-timestamp \
build/test_rendering_requires.js
@rm -rf build/slimerjs-profile
@mkdir -p build/slimerjs-profile
@cp -r test_rendering/slimerjs-profile/* build/slimerjs-profile/
node tasks/test-rendering.js
.PHONY: host-examples
host-examples: $(BUILD_HOSTED_EXAMPLES) \
$(BUILD_HOSTED)/build/ol.js \
$(BUILD_HOSTED)/build/ol-debug.js \
$(BUILD_HOSTED)/css/ol.css \
$(BUILD_HOSTED)/examples/loader.js \
$(BUILD_HOSTED)/build/ol-deps.js
.PHONY: host-libraries
host-libraries: build/timestamps/node-modules-timestamp
@rm -rf $(BUILD_HOSTED)/closure-library
@mkdir -p $(BUILD_HOSTED)/closure-library
@cp -r $(CLOSURE_LIB)/* $(BUILD_HOSTED)/closure-library/
@rm -rf $(BUILD_HOSTED)/ol/ol
@mkdir -p $(BUILD_HOSTED)/ol/ol
@cp -r src/ol/* $(BUILD_HOSTED)/ol/ol/
@rm -rf $(BUILD_HOSTED)/ol.ext
@mkdir -p $(BUILD_HOSTED)/ol.ext
@cp -r build/ol.ext/* $(BUILD_HOSTED)/ol.ext/
$(BUILD_EXAMPLES): $(EXAMPLES)
@mkdir -p $(@D)
@node tasks/build-examples.js
build/timestamps/check-%-timestamp: $(BUILD_HOSTED)/examples/%.html \
$(BUILD_HOSTED)/examples/%.js \
$(filter $(BUILD_HOSTED)/examples/resources/%,$(BUILD_HOSTED_EXAMPLES)) \
$(filter $(BUILD_HOSTED)/examples/data/%,$(BUILD_HOSTED_EXAMPLES)) \
$(BUILD_HOSTED)/examples/loader.js \
$(BUILD_HOSTED)/build/ol.js \
$(BUILD_HOSTED)/css/ol.css
@mkdir -p $(@D)
./node_modules/.bin/phantomjs --ssl-protocol=any --ignore-ssl-errors=true bin/check-example.js $(addsuffix ?mode=advanced, $<)
@touch $@
build/timestamps/check-requires-timestamp: $(SRC_JS) $(EXAMPLES_JS) \
$(SRC_SHADER_JS) $(SPEC_JS) \
$(SPEC_RENDERING JS)
@mkdir -p $(@D)
@python bin/check-requires.py $(CLOSURE_LIB) $^
@touch $@
build/timestamps/check-whitespace-timestamp: $(SRC_JS) $(EXAMPLES_JS) \
$(SPEC_JS) $(SPEC_RENDERING JS) \
$(SRC_JSDOC)
@mkdir -p $(@D)
@python bin/check-whitespace.py $^
@touch $@
build/compiled-examples/all.js: $(EXAMPLES_JS)
@mkdir -p $(@D)
@python bin/combine-examples.py $^ > $@
build/compiled-examples/all.combined.js: config/examples-all.json build/compiled-examples/all.js \
$(SRC_JS) $(SRC_SHADER_JS) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
node tasks/build.js $< $@
build/compiled-examples/%.json: config/example.json build/examples/%.js \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
@sed -e 's|{{id}}|$*|' $< > $@
build/compiled-examples/%.combined.js: build/compiled-examples/%.json \
$(SRC_JS) $(SRC_SHADER_JS) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
node tasks/build.js $< $@
build/timestamps/jsdoc-$(BRANCH)-timestamp: config/jsdoc/api/index.md \
config/jsdoc/api/conf.json $(SRC_JS) \
$(SRC_SHADER_JS) \
$(shell find config/jsdoc/api/template -type f) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
@rm -rf $(BUILD_HOSTED)/apidoc
./node_modules/.bin/jsdoc config/jsdoc/api/index.md -c config/jsdoc/api/conf.json -d $(BUILD_HOSTED)/apidoc
@touch $@
build/timestamps/gjslint-timestamp: $(SRC_JS) $(SPEC_JS) $(SPEC_RENDERING_JS) \
$(EXAMPLES_JS)
@mkdir -p $(@D)
@echo "Running gjslint..."
@gjslint --jslint_error=all --custom_jsdoc_tags=event,fires,function,classdesc,api,observable --strict $?
@touch $@
$(BUILD_HOSTED_EXAMPLES_JS): $(BUILD_HOSTED)/examples/%.js: build/examples/%.js
@mkdir -p $(@D)
@python bin/split-example.py $< $(@D)
$(BUILD_HOSTED)/examples/loader.js: bin/loader_hosted_examples.js
@mkdir -p $(@D)
@cp $< $@
$(BUILD_HOSTED)/examples/%: build/examples/%
@mkdir -p $(@D)
@cp $< $@
$(BUILD_HOSTED)/build/ol.js: build/ol.js
@mkdir -p $(@D)
@cp $< $@
$(BUILD_HOSTED)/build/ol-debug.js: build/ol-debug.js
@mkdir -p $(@D)
@cp $< $@
$(BUILD_HOSTED)/css/ol.css: build/ol.css
@mkdir -p $(@D)
@cp $< $@
$(BUILD_HOSTED)/build/ol-deps.js: host-libraries
@mkdir -p $(@D)
@python $(CLOSURE_LIB)/closure/bin/build/depswriter.py \
--root_with_prefix "src ../../../ol" \
--root_with_prefix "build/ol.ext ../../../ol.ext" \
--root $(BUILD_HOSTED)/closure-library/closure/goog \
--root_with_prefix "$(BUILD_HOSTED)/closure-library/third_party ../../third_party" \
--output_file $@
build/timestamps/jshint-timestamp: $(SRC_JS) $(SPEC_JS) $(SPEC_RENDERING_JS) \
$(TASKS_JS) $(EXAMPLES_JS) \
examples/resources/common.js \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
@echo "Running jshint..."
@./node_modules/.bin/jshint --verbose $?
@touch $@
build/timestamps/node-modules-timestamp: package.json
@mkdir -p $(@D)
npm install
@touch $@
build/ol.css: css/ol.css build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
@echo "Running cleancss..."
@./node_modules/.bin/cleancss $< > $@
build/ol.js: config/ol.json $(SRC_JS) $(SRC_SHADER_JS) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
node tasks/build.js $< $@
@$(STAT_UNCOMPRESSED) $@
@cp $@ /tmp/
@gzip /tmp/ol.js
@$(STAT_COMPRESSED) /tmp/ol.js.gz
@rm /tmp/ol.js.gz
build/ol.js.map: config/ol.json $(SRC_JS) $(SRC_SHADER_JS) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
node tasks/build.js $< $@
build/ol-debug.js: config/ol-debug.json $(SRC_JS) $(SRC_SHADER_JS) \
build/timestamps/node-modules-timestamp
@mkdir -p $(@D)
node tasks/build.js $< $@
@$(STAT_UNCOMPRESSED) $@
@cp $@ /tmp/
@gzip /tmp/ol-debug.js
@$(STAT_COMPRESSED) /tmp/ol-debug.js.gz
@rm /tmp/ol-debug.js.gz
build/test_requires.js: $(SPEC_JS) $(SRC_JS)
@mkdir -p $(@D)
@node tasks/generate-requires.js $^ > $@
build/test_rendering_requires.js: $(SPEC_RENDERING_JS)
@mkdir -p $(@D)
@node tasks/generate-requires.js $^ > $@
%shader.js: %.glsl src/ol/webgl/shader.mustache bin/pyglslunit.py
@python bin/pyglslunit.py --input $< --template src/ol/webgl/shader.mustache --output $@

190
bin/check-requires.py Normal file
View File

@@ -0,0 +1,190 @@
import os
import logging
import re
import sys
logging.basicConfig(format='%(asctime)s %(name)s: %(message)s',
level=logging.INFO)
logger = logging.getLogger('check-requires')
class Node(object):
def __init__(self):
self.present = False
self.children = {}
def _build_re(self, key):
if key == '*':
assert len(self.children) == 0
# We want to match `.doIt` but not `.SomeClass` or `.more.stuff`
return '(?=\\.[a-z]\\w*\\b(?!\\.))'
elif len(self.children) == 1:
child_key, child = next(self.children.iteritems())
child_re = child._build_re(child_key)
if child_key != '*':
child_re = '\\.' + child_re
if self.present:
return key + '(' + child_re + ')?'
else:
return key + child_re
elif self.children:
children_re = '(?:' + '|'.join(
('\\.' if k != '*' else '') + self.children[k]._build_re(k)
for k in sorted(self.children.keys())) + ')'
if self.present:
return key + children_re + '?'
else:
return key + children_re
else:
assert self.present
return key
def build_re(self, key):
return re.compile('\\b' + self._build_re(key) + '\\b')
def ifind(*paths):
"""ifind is an iterative version of os.walk, yielding all walked paths and
normalizing paths to use forward slashes."""
for path in paths:
for dirpath, dirnames, names in os.walk(path):
for name in names:
if os.sep == '/':
yield os.path.join(dirpath, name)
else:
yield '/'.join(dirpath.split(os.sep) + [name])
def _strip_comments(lines):
# FIXME this is a horribe hack, we should use a proper JavaScript parser
# here
in_multiline_comment = False
lineno = 0
for line in lines:
lineno += 1
if in_multiline_comment:
index = line.find('*/')
if index != -1:
in_multiline_comment = False
line = line[index + 2:]
if not in_multiline_comment:
line = re.sub(r'//[^\n]*', '', line)
line = re.sub(r'/\*.*?\*/', '', line)
index = line.find('/*')
if index != -1:
yield lineno, line[:index]
in_multiline_comment = True
else:
yield lineno, line
def check_requires(closure_lib, *filenames):
unused_count = 0
all_provides = set()
for filename in ifind(closure_lib):
if filename.endswith('.js'):
if not re.match(r'.*/closure/goog/', filename):
continue
# Skip goog.i18n because it contains so many modules that it causes
# the generated regular expression to exceed Python's limits
if re.match(r'.*/closure/goog/i18n/', filename):
continue
for line in open(filename, 'rU'):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
all_provides.add(m.group(1))
for filename in sorted(filenames):
require_linenos = {}
uses = set()
lines = open(filename, 'rU').readlines()
for lineno, line in _strip_comments(lines):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
all_provides.add(m.group(1))
continue
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
require_linenos[m.group(1)] = lineno
continue
ignore_linenos = require_linenos.values()
for lineno, line in enumerate(lines):
if lineno in ignore_linenos:
continue
for require in require_linenos.iterkeys():
if require in line:
uses.add(require)
for require in sorted(set(require_linenos.keys()) - uses):
logger.info('%s:%d: unused goog.require: %r' % (
filename, require_linenos[require], require))
unused_count += 1
all_provides.discard('ol')
all_provides.discard('ol.MapProperty')
root = Node()
for provide in all_provides:
node = root
for component in provide.split('.'):
if component not in node.children:
node.children[component] = Node()
node = node.children[component]
if component[0].islower():
# We've arrived at a namespace provide like `ol.foo`.
# In this case, we want to match uses like `ol.foo.doIt()` but
# not match things like `new ol.foo.SomeClass()`.
# For this purpose, we use the special wildcard key for the child.
node.children['*'] = Node()
else:
node.present = True
provide_res = [child.build_re(key)
for key, child in root.children.iteritems()]
missing_count = 0
for filename in sorted(filenames):
provides = set()
requires = set()
uses = set()
uses_linenos = {}
for lineno, line in _strip_comments(open(filename, 'rU')):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
provides.add(m.group(1))
continue
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
requires.add(m.group(1))
continue
while True:
for provide_re in provide_res:
m = provide_re.search(line)
if m:
uses.add(m.group())
uses_linenos[m.group()] = lineno
line = line[:m.start()] + line[m.end():]
break
else:
break
if filename == 'src/ol/renderer/layerrenderer.js':
uses.discard('ol.renderer.Map')
m = re.match(
r'src/ol/renderer/(\w+)/\1(\w*)layerrenderer\.js\Z', filename)
if m:
uses.discard('ol.renderer.Map')
uses.discard('ol.renderer.%s.Map' % (m.group(1),))
missing_requires = uses - requires - provides
if missing_requires:
for missing_require in sorted(missing_requires):
logger.info("%s:%d missing goog.require('%s')" %
(filename, uses_linenos[missing_require],
missing_require))
missing_count += 1
if unused_count or missing_count:
logger.error('%d unused goog.requires, %d missing goog.requires' %
(unused_count, missing_count))
if __name__ == "__main__":
check_requires(*sys.argv[1:])

44
bin/check-whitespace.py Normal file
View File

@@ -0,0 +1,44 @@
import logging
import re
import sys
logging.basicConfig(format='%(asctime)s %(name)s: %(message)s',
level=logging.INFO)
logger = logging.getLogger('check-whitespace')
CR_RE = re.compile(r'\r')
LEADING_WHITESPACE_RE = re.compile(r'\s+')
TRAILING_WHITESPACE_RE = re.compile(r'\s+\n\Z')
NO_NEWLINE_RE = re.compile(r'[^\n]\Z')
ALL_WHITESPACE_RE = re.compile(r'\s+\Z')
def check_whitespace(*filenames):
errors = 0
for filename in sorted(filenames):
whitespace = False
for lineno, line in enumerate(open(filename, 'rU')):
if lineno == 0 and LEADING_WHITESPACE_RE.match(line):
logger.info('%s:%d: leading whitespace', filename, lineno + 1)
errors += 1
if CR_RE.search(line):
logger.info('%s:%d: carriage return character in line',
filename, lineno + 1)
errors += 1
if TRAILING_WHITESPACE_RE.search(line):
logger.info('%s:%d: trailing whitespace', filename, lineno + 1)
errors += 1
if NO_NEWLINE_RE.search(line):
logger.info('%s:%d: no newline at end of file', filename,
lineno + 1)
errors += 1
whitespace = ALL_WHITESPACE_RE.match(line)
if whitespace:
logger.info('%s: trailing whitespace at end of file', filename)
errors += 1
if errors:
logger.error('%d whitespace errors' % (errors,))
if __name__ == "__main__":
check_whitespace(*sys.argv[1:])

View File

@@ -6,8 +6,8 @@
* loads Closure Library's base.js, ol-deps.js, the example's "goog.require"
* script, and the example's script in "development" mode.
*
* The ol.js and ol-deps.js scripts are built by OL3's build.py script.
* They are located in the ../build/ directory, relative to this script.
* The ol.js and ol-deps.js scripts are built using ol3's Makefile. They are
* located in the ../build/ directory, relative to this script.
*
* The script must be named loader.js.
*

39
bin/split-example.py Normal file
View File

@@ -0,0 +1,39 @@
import os
import re
import sys
def split_example_file(example, dst_dir):
lines = open(example, 'rU').readlines()
target_lines = []
target_require_lines = []
found_requires = False
found_code = False
for line in lines:
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
found_requires = True
target_require_lines.append(line)
elif found_requires:
if found_code or line not in ('\n', '\r\n'):
found_code = True
target_lines.append(line)
target = open(
os.path.join(dst_dir, os.path.basename(example)), 'wb')
target_require = open(
os.path.join(dst_dir, os.path.basename(example)
.replace('.js', '-require.js')),
'wb')
target.writelines(target_lines)
target.close()
target_require.writelines(target_require_lines)
target_require.close()
if __name__ == '__main__':
split_example_file(*sys.argv[1:])

View File

@@ -1,2 +0,0 @@
@echo off
%SystemDrive%\Python27\python.exe build.py %*

831
build.py
View File

@@ -1,831 +0,0 @@
#!/usr/bin/env python
from cStringIO import StringIO
import glob
import gzip
import json
import multiprocessing
import os
import re
import signal
import shutil
import sys
from pake import Target
from pake import ifind, main, output, rule, target, variables, virtual, which
from Queue import Queue
from threading import Thread
def sigint_handler(signal, frame):
print('Exiting')
sys.exit(0)
class ThreadPool:
"""A basic pool of worker threads"""
class Worker(Thread):
def __init__(self, tasks):
Thread.__init__(self)
self.tasks = tasks
self.daemon = True # threads will be killed on exit
self.start()
def run(self):
while True:
# block until a task is ready to be done
function, args, kargs = self.tasks.get()
try:
function(*args, **kargs)
except:
print("ERROR")
for count, thing in enumerate(args):
print '{0}. {1}'.format(count, thing)
print(sys.exc_info()[0])
print("ERROR")
self.tasks.errors = True
self.tasks.task_done()
def __init__(self, num_threads = multiprocessing.cpu_count() + 1):
self.tasks = Queue(num_threads)
self.tasks.errors = False
# create num_threads Workers, by default the number of CPUs + 1
for _ in range(num_threads): self.Worker(self.tasks)
def add_task(self, function, *args, **kargs):
self.tasks.put((function, args, kargs))
def wait_completion(self):
# wait for the queue to be empty
self.tasks.join()
return self.tasks.errors
if sys.platform == 'win32':
win = {
'CLEANCSS': './node_modules/.bin/cleancss',
'GIT': 'git.exe',
'GJSLINT': 'gjslint.exe',
'JSDOC': './node_modules/.bin/jsdoc',
'JSHINT': './node_modules/.bin/jshint',
'PYTHON': 'python.exe',
'PHANTOMJS': './node_modules/.bin/phantomjs'
}
sys_dir = os.environ.get('SYSTEMDRIVE')
program_files = os.environ.get('PROGRAMFILES')
if not which(win['GIT']):
win['GIT'] = os.path.join(program_files, 'Git', 'cmd', 'git.exe')
if not which(win['GIT']):
win['GIT'] = os.path.join(program_files, 'Git', 'bin', 'git.exe')
if not which(win['PYTHON']):
win['PYTHON'] = os.path.join(sys_dir, 'Python27', 'python.exe')
if not which(win['GJSLINT']):
win['GJSLINT'] = os.path.join(sys_dir, 'Python27', 'Scripts', 'gjslint.exe')
if not which(win['PHANTOMJS']):
win['PHANTOMJS'] = 'phantomjs.exe'
if not which(win['PHANTOMJS']):
win['PHANTOMJS'] = os.path.join(sys_dir, 'phantomjs-1.9.7-windows', 'phantomjs.exe')
if not which(win['JSDOC']):
win['JSDOC'] = os.path.join(program_files, 'jsdoc3', 'jsdoc.cmd')
for program, path in win.iteritems():
setattr(variables, program, path)
else:
variables.CLEANCSS = './node_modules/.bin/cleancss'
variables.GIT = 'git'
variables.GJSLINT = 'gjslint'
variables.JSHINT = './node_modules/.bin/jshint'
variables.JSDOC = './node_modules/.bin/jsdoc'
variables.PYTHON = 'python'
variables.PHANTOMJS = './node_modules/.bin/phantomjs'
variables.BRANCH = output(
'%(GIT)s', 'rev-parse', '--abbrev-ref', 'HEAD').strip()
EXECUTABLES = [variables.CLEANCSS, variables.GIT, variables.GJSLINT,
variables.JSDOC, variables.JSHINT, variables.PYTHON,
variables.PHANTOMJS]
EXAMPLES_SRC_ALL = [path for path in ifind('examples')]
EXAMPLES_SRC_HTML = [path
for path in EXAMPLES_SRC_ALL
if path.endswith('.html')
if path != 'examples/index.html']
EXAMPLES_SRC_JS = [example.replace('.html', '.js')
for example in EXAMPLES_SRC_HTML]
EXAMPLES_DEST_ALL = [path.replace('examples', 'build/examples')
for path in EXAMPLES_SRC_ALL]
GLSL_SRC = [path
for path in ifind('src')
if path.endswith('.glsl')]
JSDOC_SRC = [path
for path in ifind('src')
if path.endswith('.jsdoc')]
SHADER_SRC = [path.replace('.glsl', 'shader.js')
for path in GLSL_SRC]
SPEC = [path
for path in ifind('test/spec')
if path.endswith('.js')]
SPEC_RENDERING = [path
for path in ifind('test_rendering/spec')
if path.endswith('.js')]
TASKS = [path
for path in ifind('tasks')
if path.endswith('.js')]
SRC = [path
for path in ifind('src/ol')
if path.endswith('.js')
if path not in SHADER_SRC]
NPM_INSTALL = 'build/npm-install-timestamp'
def report_sizes(t):
stringio = StringIO()
gzipfile = gzip.GzipFile(t.name, 'w', 9, stringio)
with open(t.name, 'rb') as f:
shutil.copyfileobj(f, gzipfile)
gzipfile.close()
rawsize = os.stat(t.name).st_size
gzipsize = len(stringio.getvalue())
savings = '{0:.2%}'.format((rawsize - gzipsize)/float(rawsize))
t.info('uncompressed: %8d bytes', rawsize)
t.info(' compressed: %8d bytes, (saved %s)', gzipsize, savings)
virtual('default', 'build')
virtual('ci', 'lint', 'build', 'test', 'test-rendering',
'build/compiled-examples/all.combined.js', 'check-examples', 'apidoc')
virtual('build', 'build/ol.css', 'build/ol.js', 'build/ol-debug.js',
'build/ol.js.map')
virtual('check', 'lint', 'build/ol.js', 'test')
virtual('todo', 'fixme')
@target(NPM_INSTALL, 'package.json')
def npm_install(t):
t.run('npm', 'install')
t.touch()
@target('build/ol.css', 'css/ol.css', NPM_INSTALL)
def build_ol_css(t):
t.output('%(CLEANCSS)s', 'css/ol.css')
def _build_js(t):
t.run('node', 'tasks/build.js', 'config/ol.json', 'build/ol.js')
@target('build/ol.js', SRC, SHADER_SRC, 'config/ol.json', NPM_INSTALL)
def build_ol_js(t):
_build_js(t)
report_sizes(t)
@target('build/ol.js.map', SRC, SHADER_SRC, 'config/ol.json', NPM_INSTALL)
def build_ol_js_map(t):
_build_js(t)
@target('build/ol-debug.js', SRC, SHADER_SRC, 'config/ol-debug.json',
NPM_INSTALL)
def build_ol_debug_js(t):
t.run('node', 'tasks/build.js', 'config/ol-debug.json', 'build/ol-debug.js')
report_sizes(t)
for glsl_src in GLSL_SRC:
def shader_src_helper(glsl_src):
@target(glsl_src.replace('.glsl', 'shader.js'), glsl_src,
'src/ol/webgl/shader.mustache', 'bin/pyglslunit.py')
def shader_src(t):
t.run('%(PYTHON)s', 'bin/pyglslunit.py',
'--input', glsl_src,
'--template', 'src/ol/webgl/shader.mustache',
'--output', t.name)
shader_src_helper(glsl_src)
def build_requires(task):
requires = set()
for dependency in task.dependencies:
for line in open(dependency, 'rU'):
match = re.match(r'goog\.provide\(\'(.*)\'\);', line)
if match:
requires.add(match.group(1))
with open(task.name, 'wb') as f:
for require in sorted(requires):
f.write('goog.require(\'%s\');\n' % (require,))
@target('build/test_requires.js', SPEC)
def build_test_requires(t):
build_requires(t)
@target('build/test_rendering_requires.js', SPEC_RENDERING)
def build_test_rendering_requires(t):
build_requires(t)
virtual('examples', EXAMPLES_DEST_ALL)
@rule(r'\Abuild\/examples/(?P<filepath>.*)\Z')
def examples_dest(name, match):
def action(t):
t.run('node', 'tasks/build-examples.js')
dependencies = ['examples/%(filepath)s' % match.groupdict()]
return Target(name, action=action, dependencies=dependencies)
@target('build/compiled-examples/all.combined.js', 'build/compiled-examples/all.js',
SRC, SHADER_SRC, 'config/examples-all.json', NPM_INSTALL)
def build_examples_all_combined_js(t):
t.run('node', 'tasks/build.js', 'config/examples-all.json',
'build/compiled-examples/all.combined.js')
report_sizes(t)
@target('build/compiled-examples/all.js', EXAMPLES_SRC_JS)
def build_examples_all_js(t):
t.output('%(PYTHON)s', 'bin/combine-examples.py', t.dependencies)
@rule(r'\Abuild/compiled-examples/(?P<id>.*).json\Z')
def examples_star_json(name, match):
def action(t):
# When compiling the ol3 code and the application code together it is
# better to use oli.js and olx.js files as "input" files rather than
# "externs" files. Indeed, externs prevent renaming, which is neither
# necessary nor desirable in this case.
#
# oli.js and olx.js do not provide or require namespaces (using
# "goog.provide" or "goog.require"). For that reason, if they are
# specified as input files through the "src" property, then
# closure-util will exclude them when creating the dependencies graph.
# So the compile "js" property is used instead. With that property the
# oli.js and olx.js files are passed directly to the compiler. And by
# setting "manage_closure_dependencies" to "true" the compiler will not
# exclude them from its dependencies graph.
content = json.dumps({
"exports": [],
"src": [
"src/**/*.js",
"build/ol.ext/*.js",
"build/examples/%(id)s.js" % match.groupdict()],
"compile": {
"js": [
"externs/olx.js",
"externs/oli.js",
],
"externs": [
"externs/bingmaps.js",
"externs/bootstrap.js",
"externs/closure-compiler.js",
"externs/esrijson.js",
"externs/example.js",
"externs/fastclick.js",
"externs/geojson.js",
"externs/jquery-1.9.js",
"externs/proj4js.js",
"externs/tilejson.js",
"externs/topojson.js"
],
"define": [
"goog.array.ASSUME_NATIVE_FUNCTIONS=true",
"goog.dom.ASSUME_STANDARDS_MODE=true",
"goog.json.USE_NATIVE_JSON=true",
"goog.DEBUG=false"
],
"jscomp_error": [
"accessControls",
"ambiguousFunctionDecl",
"checkDebuggerStatement",
"checkEventfulObjectDisposal",
"checkRegExp",
"checkStructDictInheritance",
"checkTypes",
"checkVars",
"const",
"constantProperty",
"deprecated",
"duplicate",
"duplicateMessage",
"es3",
"es5Strict",
"externsValidation",
"fileoverviewTags",
"globalThis",
"internetExplorerChecks",
"invalidCasts",
"misplacedTypeAnnotation",
"missingProperties",
"missingProvide",
"missingRequire",
"missingReturn",
"newCheckTypes",
"nonStandardJsDocs",
"strictModuleDepCheck",
"suspiciousCode",
"typeInvalidation",
"tweakValidation",
"undefinedNames",
"undefinedVars",
"uselessCode",
"violatedModuleDep",
"visibility"
],
"jscomp_off": [
"unknownDefines"
],
"extra_annotation_name": [
"api", "observable"
],
"compilation_level": "ADVANCED",
"warning_level": "VERBOSE",
"output_wrapper": "(function(){%output%})();",
"use_types_for_optimization": True,
"manage_closure_dependencies": True
}
})
with open(t.name, 'wb') as f:
f.write(content)
return Target(name, action=action,
dependencies=[__file__, NPM_INSTALL])
@rule(r'\Abuild/compiled-examples/(?P<id>.*).combined.js\Z')
def examples_star_combined_js(name, match):
def action(t):
config = 'build/compiled-examples/%(id)s.json' % match.groupdict()
t.run('node', 'tasks/build.js', config, name)
report_sizes(t)
dependencies = [SRC, SHADER_SRC,
'examples/%(id)s.js' % match.groupdict(),
'build/compiled-examples/%(id)s.json' % match.groupdict(),
NPM_INSTALL]
return Target(name, action=action, dependencies=dependencies)
@target('serve', 'build/test_requires.js', 'build/test_rendering_requires.js',
NPM_INSTALL)
def serve(t):
t.run('node', 'tasks/serve.js')
virtual('lint', 'build/lint-timestamp', 'build/check-requires-timestamp',
'build/check-whitespace-timestamp', 'jshint')
@target('build/lint-timestamp', SRC, EXAMPLES_SRC_JS, SPEC, SPEC_RENDERING,
precious=True)
def build_lint_src_timestamp(t):
t.run('%(GJSLINT)s',
'--jslint_error=all',
'--custom_jsdoc_tags=event,fires,function,classdesc,api,observable',
'--strict',
t.newer(t.dependencies))
t.touch()
virtual('jshint', 'build/jshint-timestamp')
@target('build/jshint-timestamp', SRC, EXAMPLES_SRC_JS, SPEC, SPEC_RENDERING,
'examples/resources/common.js', TASKS, NPM_INSTALL, precious=True)
def build_jshint_timestamp(t):
t.run(variables.JSHINT, '--verbose', t.newer(t.dependencies))
t.touch()
def _strip_comments(lines):
# FIXME this is a horribe hack, we should use a proper JavaScript parser
# here
in_multiline_comment = False
lineno = 0
for line in lines:
lineno += 1
if in_multiline_comment:
index = line.find('*/')
if index != -1:
in_multiline_comment = False
line = line[index + 2:]
if not in_multiline_comment:
line = re.sub(r'//[^\n]*', '', line)
line = re.sub(r'/\*.*?\*/', '', line)
index = line.find('/*')
if index != -1:
yield lineno, line[:index]
in_multiline_comment = True
else:
yield lineno, line
@target('build/check-requires-timestamp', SRC, EXAMPLES_SRC_JS, SHADER_SRC,
SPEC, SPEC_RENDERING)
def build_check_requires_timestamp(t):
unused_count = 0
all_provides = set()
closure_lib_path = output('node', '-e',
'process.stdout.write(require("closure-util").getLibraryPath())')
for filename in ifind(closure_lib_path):
if filename.endswith('.js'):
if not re.match(r'.*/closure/goog/', filename):
continue
# Skip goog.i18n because it contains so many modules that it causes
# the generated regular expression to exceed Python's limits
if re.match(r'.*/closure/goog/i18n/', filename):
continue
for line in open(filename, 'rU'):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
all_provides.add(m.group(1))
for filename in sorted(t.dependencies):
require_linenos = {}
uses = set()
lines = open(filename, 'rU').readlines()
for lineno, line in _strip_comments(lines):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
all_provides.add(m.group(1))
continue
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
require_linenos[m.group(1)] = lineno
continue
ignore_linenos = require_linenos.values()
for lineno, line in enumerate(lines):
if lineno in ignore_linenos:
continue
for require in require_linenos.iterkeys():
if require in line:
uses.add(require)
for require in sorted(set(require_linenos.keys()) - uses):
t.info('%s:%d: unused goog.require: %r' % (
filename, require_linenos[require], require))
unused_count += 1
all_provides.discard('ol')
all_provides.discard('ol.MapProperty')
class Node(object):
def __init__(self):
self.present = False
self.children = {}
def _build_re(self, key):
if key == '*':
assert len(self.children) == 0
# We want to match `.doIt` but not `.SomeClass` or `.more.stuff`
return '(?=\\.[a-z]\\w*\\b(?!\\.))'
elif len(self.children) == 1:
child_key, child = next(self.children.iteritems())
child_re = child._build_re(child_key)
if child_key != '*':
child_re = '\\.' + child_re
if self.present:
return key + '(' + child_re + ')?'
else:
return key + child_re
elif self.children:
children_re = '(?:' + '|'.join(
('\\.' if k != '*' else '') + self.children[k]._build_re(k)
for k in sorted(self.children.keys())) + ')'
if self.present:
return key + children_re + '?'
else:
return key + children_re
else:
assert self.present
return key
def build_re(self, key):
return re.compile('\\b' + self._build_re(key) + '\\b')
root = Node()
for provide in all_provides:
node = root
for component in provide.split('.'):
if component not in node.children:
node.children[component] = Node()
node = node.children[component]
if component[0].islower():
# We've arrived at a namespace provide like `ol.foo`.
# In this case, we want to match uses like `ol.foo.doIt()` but
# not match things like `new ol.foo.SomeClass()`.
# For this purpose, we use the special wildcard key for the child.
node.children['*'] = Node()
else:
node.present = True
provide_res = [child.build_re(key)
for key, child in root.children.iteritems()]
missing_count = 0
for filename in sorted(t.dependencies):
provides = set()
requires = set()
uses = set()
uses_linenos = {}
for lineno, line in _strip_comments(open(filename, 'rU')):
m = re.match(r'goog.provide\(\'(.*)\'\);', line)
if m:
provides.add(m.group(1))
continue
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
requires.add(m.group(1))
continue
while True:
for provide_re in provide_res:
m = provide_re.search(line)
if m:
uses.add(m.group())
uses_linenos[m.group()] = lineno
line = line[:m.start()] + line[m.end():]
break
else:
break
if filename == 'src/ol/renderer/layerrenderer.js':
uses.discard('ol.renderer.Map')
m = re.match(
r'src/ol/renderer/(\w+)/\1(\w*)layerrenderer\.js\Z', filename)
if m:
uses.discard('ol.renderer.Map')
uses.discard('ol.renderer.%s.Map' % (m.group(1),))
missing_requires = uses - requires - provides
if missing_requires:
for missing_require in sorted(missing_requires):
t.info("%s:%d missing goog.require('%s')" %
(filename, uses_linenos[missing_require], missing_require))
missing_count += 1
if unused_count or missing_count:
t.error('%d unused goog.requires, %d missing goog.requires' %
(unused_count, missing_count))
t.touch()
@target('build/check-whitespace-timestamp', SRC, EXAMPLES_SRC_JS,
SPEC, SPEC_RENDERING, JSDOC_SRC, precious=True)
def build_check_whitespace_timestamp(t):
CR_RE = re.compile(r'\r')
LEADING_WHITESPACE_RE = re.compile(r'\s+')
TRAILING_WHITESPACE_RE = re.compile(r'\s+\n\Z')
NO_NEWLINE_RE = re.compile(r'[^\n]\Z')
ALL_WHITESPACE_RE = re.compile(r'\s+\Z')
errors = 0
for filename in sorted(t.newer(t.dependencies)):
whitespace = False
for lineno, line in enumerate(open(filename, 'rU')):
if lineno == 0 and LEADING_WHITESPACE_RE.match(line):
t.info('%s:%d: leading whitespace', filename, lineno + 1)
errors += 1
if CR_RE.search(line):
t.info('%s:%d: carriage return character in line', filename, lineno + 1)
errors += 1
if TRAILING_WHITESPACE_RE.search(line):
t.info('%s:%d: trailing whitespace', filename, lineno + 1)
errors += 1
if NO_NEWLINE_RE.search(line):
t.info('%s:%d: no newline at end of file', filename, lineno + 1)
errors += 1
whitespace = ALL_WHITESPACE_RE.match(line)
if whitespace:
t.info('%s: trailing whitespace at end of file', filename)
errors += 1
if errors:
t.error('%d whitespace errors' % (errors,))
t.touch()
virtual('apidoc', 'build/jsdoc-%(BRANCH)s-timestamp' % vars(variables))
@target('build/jsdoc-%(BRANCH)s-timestamp' % vars(variables),
SRC, SHADER_SRC, ifind('config/jsdoc/api/template'),
NPM_INSTALL)
def jsdoc_BRANCH_timestamp(t):
t.run('%(JSDOC)s', 'config/jsdoc/api/index.md',
'-c', 'config/jsdoc/api/conf.json',
'-d', 'build/hosted/%(BRANCH)s/apidoc')
t.touch()
def split_example_file(example, dst_dir):
lines = open(example, 'rU').readlines()
target_lines = []
target_require_lines = []
found_requires = False
found_code = False
for line in lines:
m = re.match(r'goog.require\(\'(.*)\'\);', line)
if m:
found_requires = True
target_require_lines.append(line)
elif found_requires:
if found_code or line not in ('\n', '\r\n'):
found_code = True
target_lines.append(line)
target = open(
os.path.join(dst_dir, os.path.basename(example)), 'wb')
target_require = open(
os.path.join(dst_dir, os.path.basename(example)
.replace('.js', '-require.js')),
'wb')
target.writelines(target_lines)
target.close()
target_require.writelines(target_require_lines)
target_require.close()
@target('host-examples', 'build', 'examples', phony=True)
def host_examples(t):
examples_dir = 'build/hosted/%(BRANCH)s/examples'
build_dir = 'build/hosted/%(BRANCH)s/build'
css_dir = 'build/hosted/%(BRANCH)s/css'
closure_lib_path = output('node', '-e',
'process.stdout.write(require("closure-util").getLibraryPath())')
t.rm_rf(examples_dir)
t.cp_r('build/examples', examples_dir)
for example in EXAMPLES_SRC_JS:
split_example_file(example, examples_dir % vars(variables))
t.cp('bin/loader_hosted_examples.js', examples_dir + '/loader.js')
t.rm_rf(build_dir)
t.makedirs(build_dir)
t.rm_rf(css_dir)
t.makedirs(css_dir)
t.cp('build/ol.js', 'build/ol-debug.js', build_dir)
t.cp('build/ol.css', css_dir)
t.rm_rf('build/hosted/%(BRANCH)s/closure-library')
t.cp_r(closure_lib_path, 'build/hosted/%(BRANCH)s/closure-library')
t.rm_rf('build/hosted/%(BRANCH)s/ol')
t.makedirs('build/hosted/%(BRANCH)s/ol')
t.cp_r('src/ol', 'build/hosted/%(BRANCH)s/ol/ol')
t.rm_rf('build/hosted/%(BRANCH)s/ol.ext')
t.cp_r('build/ol.ext', 'build/hosted/%(BRANCH)s/ol.ext')
t.run('%(PYTHON)s', closure_lib_path + '/closure/bin/build/depswriter.py',
'--root_with_prefix', 'src ../../../ol',
'--root_with_prefix', 'build/ol.ext ../../../ol.ext',
'--root', 'build/hosted/%(BRANCH)s/closure-library/closure/goog',
'--root_with_prefix', 'build/hosted/%(BRANCH)s/closure-library/'
'third_party ../../third_party',
'--output_file', 'build/hosted/%(BRANCH)s/build/ol-deps.js')
@target('check-examples', 'host-examples', phony=True)
def check_examples(t):
examples = ['build/hosted/%(BRANCH)s/' + e
for e in EXAMPLES_SRC_HTML
if not open(e.replace('.html', '.js'), 'rU').readline().startswith('// NOCOMPILE')]
all_examples = [e + '?mode=advanced' for e in examples]
# Run the examples checks in a pool of threads
pool = ThreadPool()
for example in all_examples:
pool.add_task(t.run, '%(PHANTOMJS)s', '--ssl-protocol=any',
'--ignore-ssl-errors=true', 'bin/check-example.js', example)
errors = pool.wait_completion()
if errors:
sys.exit(1)
@target('test', NPM_INSTALL, 'build/test_requires.js', phony=True)
def test(t):
t.run('node', 'tasks/test.js')
@target('test-coverage', NPM_INSTALL, phony=True)
def test_coverage(t):
t.run('node', 'tasks/test-coverage.js')
@target('test-rendering', 'build/test_rendering_requires.js',
NPM_INSTALL, phony=True)
def test_rendering(t):
# create a temp. profile to run the tests with WebGL
tmp_profile_dir = 'build/slimerjs-profile'
t.rm_rf(tmp_profile_dir)
t.cp_r('test_rendering/slimerjs-profile', tmp_profile_dir)
t.run('node', 'tasks/test-rendering.js')
@target('fixme', phony=True)
def find_fixme(t):
regex = re.compile('FIXME|TODO')
matches = dict()
totalcount = 0
for filename in SRC:
f = open(filename, 'r')
for lineno, line in enumerate(f):
if regex.search(line):
if (filename not in matches):
matches[filename] = list()
matches[filename].append('#%-10d %s' % (
lineno + 1, line.strip()))
totalcount += 1
f.close()
for filename in matches:
num_matches = len(matches[filename])
noun = 'matches' if num_matches > 1 else 'match'
print ' %s has %d %s:' % (filename, num_matches, noun)
for match in matches[filename]:
print ' %s' % (match,)
print
print 'A total of %d TODO/FIXME(s) were found' % (totalcount,)
@target('reallyclean')
def reallyclean(t):
"""Removes untracked files and folders from previous builds."""
# -X => only clean up files that are usually ignored e.g.
# through .gitignore
# -d => also consider directories for deletion
# -f => if git configuration variable clean.requireForce != false,
# git clean will refuse to run unless given -f or -n.
t.run('%(GIT)s', 'clean', '-X', '-d', '-f', '.')
@target('checkdeps')
def check_dependencies(t):
for exe in EXECUTABLES:
status = 'present' if which(exe) else 'MISSING'
print 'Program "%s" seems to be %s.' % (exe, status)
print 'For certain targets all above programs need to be present.'
@target('help')
def display_help(t):
print '''
build.py - The OpenLayers 3 build script.
Usage:
./build.py [options] [target] (on Unix-based machines)
<python-executable.exe> build.py [options] [target] (on Windows machines)
There is one option:
-c - Cleans up the repository from previous builds.
The most common targets are:
serve - Serves files, on port 3000.
lint - Runs gjslint on all sourcefiles to enforce specific syntax.
build - Builds singlefile versions of OpenLayers JavaScript and
CSS. This is also the default build target which runs when
no target is specified.
test - Runs the testsuite and displays the results.
test-rendering - Runs the rendering testsuite and displays the results.
check - Runs the lint-target, builds some OpenLayers files, and
then runs test. Many developers call this target often
while working on the code.
help - Shows this help.
Other less frequently used targets are:
apidoc - Builds the API-Documentation using JSDoc3.
ci - Builds all examples in various modes and usually takes a
long time to finish. This target calls the following
targets: 'lint', 'build', 'test', 'test-rendering',
'build/compiled-examples/all.combined.js', 'check-examples',
and 'apidoc'. This is the target run on Travis CI.
test-coverage - Generates a test coverage report in the coverage folder.
reallyclean - Remove untracked files from the repository.
checkdeps - Checks whether all required development software is
installed on your machine.
fixme - Will print a list of parts of the code that are marked
with either TODO or FIXME.
todo - This is an alias for the fixme-target (see above).
If no target is given, the build-target will be executed.
The above list is not complete, please see the source code for not-mentioned
and only seldom called targets.
'''
if __name__ == '__main__':
signal.signal(signal.SIGINT, sigint_handler)
main()

79
config/example.json Normal file
View File

@@ -0,0 +1,79 @@
{
"exports": [],
"src": [
"src/**/*.js",
"build/ol.ext/*.js",
"build/examples/{{id}}.js"
],
"compile": {
"js": [
"externs/olx.js",
"externs/oli.js"
],
"externs": [
"externs/bingmaps.js",
"externs/bootstrap.js",
"externs/closure-compiler.js",
"externs/example.js",
"externs/fastclick.js",
"externs/geojson.js",
"externs/jquery-1.9.js",
"externs/proj4js.js",
"externs/tilejson.js",
"externs/topojson.js",
"externs/vbarray.js"
],
"define": [
"goog.array.ASSUME_NATIVE_FUNCTIONS=true",
"goog.dom.ASSUME_STANDARDS_MODE=true",
"goog.json.USE_NATIVE_JSON=true",
"goog.DEBUG=false"
],
"jscomp_error": [
"accessControls",
"ambiguousFunctionDecl",
"checkEventfulObjectDisposal",
"checkRegExp",
"checkStructDictInheritance",
"checkTypes",
"checkVars",
"const",
"constantProperty",
"deprecated",
"duplicateMessage",
"es3",
"es5Strict",
"externsValidation",
"fileoverviewTags",
"globalThis",
"internetExplorerChecks",
"invalidCasts",
"misplacedTypeAnnotation",
"missingGetCssName",
"missingProperties",
"missingProvide",
"missingRequire",
"missingReturn",
"newCheckTypes",
"nonStandardJsDocs",
"suspiciousCode",
"strictModuleDepCheck",
"typeInvalidation",
"undefinedNames",
"undefinedVars",
"uselessCode",
"visibility"
],
"jscomp_off": [
"unknownDefines"
],
"extra_annotation_name": [
"api", "observable"
],
"compilation_level": "ADVANCED",
"warning_level": "VERBOSE",
"output_wrapper": "(function(){%output%})();",
"use_types_for_optimization": true,
"manage_closure_dependencies": true
}
}

View File

@@ -12,8 +12,7 @@
"install": "node tasks/install.js",
"postinstall": "closure-util update",
"start": "node tasks/serve.js",
"test": "node tasks/test.js",
"test-coverage": "node tasks/test-coverage.js"
"test": "node tasks/test.js"
},
"main": "dist/ol.js",
"repository": {

509
pake.py
View File

@@ -1,509 +0,0 @@
#!/usr/bin/env python
import collections
import contextlib
import hashlib
import logging
import optparse
import os
import re
import shutil
import subprocess
import tempfile
import sys
import time
import urllib2
logger = logging.getLogger(__name__)
if hasattr(subprocess, 'check_output'):
check_output = subprocess.check_output
else:
# Copied with minor modifications from the Python source code
# http://hg.python.org/cpython/file/9cb1366b251b/Lib/subprocess.py#l549
def check_output(*popenargs, **kwargs):
if 'stdout' in kwargs:
raise ValueError(
'stdout argument not allowed, it will be overridden.')
process = subprocess.Popen(stdout=subprocess.PIPE,
*popenargs, **kwargs)
output, unused_err = process.communicate()
retcode = process.poll()
if retcode:
cmd = kwargs.get("args")
if cmd is None:
cmd = popenargs[0]
raise subprocess.CalledProcessError(retcode, cmd, output=output)
return output
class PakeError(RuntimeError):
pass
class AmbiguousRuleError(PakeError):
def __init__(self, name):
self.name = name
def __str__(self):
return '%r matches multiple rules' % (self.name,)
class BuildError(PakeError):
def __init__(self, target, message):
self.target = target
self.message = message
def __str__(self):
return '%s: %s' % (self.target.name, self.message)
class DuplicateTargetError(PakeError):
def __init__(self, target):
self.target = target
def __str__(self):
return 'duplicate target %r' % (self.target.name,)
class UnknownTargetError(PakeError):
def __init__(self, name):
self.name = name
def __str__(self):
return 'unknown target %r' % (self.name,)
class Target(object):
"""Target is the core object of pake. It includes all of the target's name
(which may or may not correspond to a real file in the filesystem, see the
comments in virtual and TargetCollection below), the action to be performed
when this target is to be rebuilt, its dependencies, and various other
metadata."""
def __init__(self, name, action=None, clean=True, dependencies=(),
help=None, help_group=None, makedirs=True, phony=False,
precious=False):
self.name = name
self.action = action
self._clean = clean
self.dependencies = list(flatten(dependencies))
self.help = help
self.help_group = help_group
self._makedirs = makedirs
self.phony = phony
self.precious = precious
self.logger = logging.getLogger(self.name)
self.timestamp = None
def build(self, dry_run=False):
timestamp = 0
for dependency in self.dependencies:
target = targets.get(dependency)
timestamp = max(timestamp, target.build(dry_run=dry_run))
self.debug('build')
if self.timestamp is None:
if not self.phony and os.path.exists(self.name):
self.timestamp = os.stat(self.name).st_mtime
else:
self.timestamp = -1
if self.timestamp < timestamp:
self.debug('action')
if self._makedirs and not dry_run:
self.makedirs(os.path.dirname(self.name))
if self.action:
if self.action.__doc__:
self.info(self.action.__doc__)
if not dry_run:
self.action(self)
self.timestamp = timestamp or time.time()
return self.timestamp
@contextlib.contextmanager
def chdir(self, dir):
cwd = os.getcwd()
dir = dir % vars(variables)
self.info('cd %s', dir)
os.chdir(dir)
try:
yield dir
finally:
self.info('cd %s', cwd)
os.chdir(cwd)
def cp(self, *args):
args = flatten_expand_list(args)
dest = args.pop()
for arg in args:
self.info('cp %s %s', arg, dest)
shutil.copy(arg, dest)
def cp_r(self, *args):
args = flatten_expand_list(args)
dest = args.pop()
for arg in args:
self.info('cp -r %s %s', arg, dest)
shutil.copytree(arg, dest)
def clean(self, really=False, recurse=True):
if (self._clean or really) and not self.precious:
self.info('clean')
try:
os.remove(self.name)
except OSError:
pass
if recurse:
for dependency in self.dependencies:
targets.get(dependency).clean(really=really, recurse=recurse)
def debug(self, *args, **kwargs):
self.logger.debug(*args, **kwargs)
def download(self, url, md5=None, sha1=None):
content = urllib2.urlopen(url).read()
if md5 and hashlib.md5(content).hexdigest() != md5:
raise BuildError(self, 'corrupt download')
if sha1 and hashlib.sha1(content).hexdigest() != sha1:
raise BuildError(self, 'corrupt download')
with open(self.name, 'wb') as f:
f.write(content)
def error(self, message):
raise BuildError(self, message)
def graph(self, f, visited):
if self in visited:
return
visited.add(self)
for dependency in self.dependencies:
target = targets.get(dependency)
f.write('\t"%s" -> "%s";\n' % (self.name, target.name))
target.graph(f, visited)
def info(self, *args, **kwargs):
self.logger.info(*args, **kwargs)
def makedirs(self, path):
path = path % vars(variables)
if path and not os.path.exists(path):
self.info('mkdir -p %s', path)
os.makedirs(path)
def newer(self, *args):
args = flatten_expand_list(args)
return [arg for arg in args
if targets.get(arg).timestamp > self.timestamp]
def output(self, *args, **kwargs):
"""output runs the command passed to it, saving the output of the
command to the contents of the target. For example:
@target('ofile')
def ofile(t):
t.output('echo', '123')
After this target's action is executed, ofile will contain the string
"123"."""
args = flatten_expand_list(args)
self.info(' '.join(args))
try:
output = check_output(args, **kwargs)
with open(self.name, 'wb') as f:
f.write(output)
except subprocess.CalledProcessError as e:
self.clean(recurse=False)
self.error(e)
def rm_rf(self, *args):
"""rm_rf recursively deletes the files and/or directories passed to
it."""
args = flatten_expand_list(args)
for arg in args:
self.info('rm -rf %s', arg)
shutil.rmtree(arg, ignore_errors=True)
def run(self, *args, **kwargs):
args = flatten_expand_list(args)
self.info(' '.join(args))
try:
subprocess.check_call(args, **kwargs)
except subprocess.CalledProcessError as e:
self.clean(recurse=False)
self.error(e)
@contextlib.contextmanager
def tempdir(self):
"""tempdir creates a temporary directory, changes to it, and runs the
nested block of code. However the nested block of code exits, tempdir
will delete the temporary directory permanently, before pake exits. For
example:
with t.tempdir():
# copy various files to $PWD (the temporary directory)
# zip up the contents of $PWD, or copy them somewhere else
However the above code exits (e.g. copy error or zip error), the
temporary directory will be cleaned up."""
tempdir = tempfile.mkdtemp()
self.info('mkdir -p %s', tempdir)
try:
yield tempdir
finally:
self.info('rm -rf %s', tempdir)
shutil.rmtree(tempdir, ignore_errors=True)
def touch(self):
"""touch updates the timestamp of the target. If the target already
exists as a file in the filesystem its timestamp is updated, otherwise
a new file is created with the current timestamp."""
if os.path.exists(self.name):
os.utime(self.name, None)
else:
with open(self.name, 'wb'):
pass
class TargetCollection(object):
"""TargetCollection implements a namespace for looking up build targets.
TargetCollection will first look for rules that match exactly, and then
- if no match is found - search through a list of regular expression-based
rules. As soon as a regular expression match is found, that rule is added
to the list of rules that match exactly. Typically, an invocation of pake
will only create a single TargetCollection."""
def __init__(self):
self.default = None
self.targets = {}
def add(self, target):
"""add adds a concrete target to self, raising an error if the target
already exists. If target is the first target to be added, it becomes
the default for this TargetCollection."""
if target.name in self.targets:
raise DuplicateTargetError(target)
self.targets[target.name] = target
if self.default is None:
self.default = target
def get(self, name):
"""get searches for a target. If it already exists, it is returned.
Otherwise, get searches through the defined rules, trying to find a
rule that matches. If it finds a matching rule, a concrete target is
instantiated, cached, and returned. If no match is found, a virtual
precious target is instantiated and returned."""
if name in self.targets:
return self.targets[name]
target = None
for regexp, f in rules.iteritems():
match = regexp.search(name)
if not match:
continue
if target is not None:
raise AmbiguousRuleError(name)
target = f(name, match)
if target is None:
if os.path.exists(name):
target = Target(name, precious=True)
else:
raise UnknownTargetError(name)
self.targets[name] = target
return target
def format_epilog(self, formatter):
helps_by_help_group = collections.defaultdict(dict)
max_name_len = 0
for name in sorted(self.targets):
target = self.targets[name]
if target.help is not None:
helps_by_help_group[target.help_group][name] = target.help
max_name_len = max(max_name_len, len(name))
lines = []
lines.append('Targets:\n')
format = ' %%-%ds %%s\n' % (max_name_len,)
for help_group in sorted(helps_by_help_group.keys()):
helps = helps_by_help_group[help_group]
if help_group is not None:
lines.append('%s targets:\n' % (help_group,))
for name in sorted(helps.keys()):
lines.append(format % (name, helps[name]))
return ''.join(lines)
class VariableCollection(object):
"""VariableCollection implements an object with properties where the first
set of a property wins, and all further sets are ignored. For example:
vc = VariableCollection()
vc.FOO = 1 # First set of the property FOO
vc.FOO = 2 # Further sets of the property FOO are ignored, and do
# not raise an error. After this statement, vc.FOO is
# still 1.
print vc.FOO # Prints "1" """
def __init__(self, **kwargs):
for key, value in kwargs.iteritems():
setattr(self, key, value)
def __setattr__(self, key, value):
"""Only set an attribute if it has not already been set. First to set
the value is the winner."""
if not hasattr(self, key):
object.__setattr__(self, key, value)
# targets is the single TargetCollection instance created for this invokation
# of pake
targets = TargetCollection()
# rules is a dict of regular expressions to @rules where dynamically created
# rules are registered.
rules = {}
# variables is the global set of substitution variables, where the first setter
# takes priority. The priority order is:
# 1. Environment variables
# 2. Command line arguments
# 3. Internal Python settings in build.py
variables = VariableCollection(**os.environ)
def flatten(*args):
"""flatten takes a variable number of arguments, each of which may or may
be not be a collection.Iterable, and yields the elements of each in
depth-first order. In short, it flattens nested iterables into a single
collection. For example, flatten(1, [2, (3, 4), 5], 6) yields 1, 2, 3, 4,
5, 6."""
for arg in args:
if (isinstance(arg, collections.Iterable) and
not isinstance(arg, basestring)):
for element in flatten(*arg):
yield element
else:
yield arg
def flatten_expand_list(*args):
"""flatten_expand_list applies flatten, treats each element as a string,
and formats each string according to the global value of variables."""
return list(arg % vars(variables) for arg in flatten(args))
def ifind(*paths):
"""ifind is an iterative version of os.walk, yielding all walked paths and
normalizing paths to use forward slashes."""
for path in paths:
for dirpath, dirnames, names in os.walk(path):
for name in names:
if os.sep == '/':
yield os.path.join(dirpath, name)
else:
yield '/'.join(dirpath.split(os.sep) + [name])
def main(argv=sys.argv):
option_parser = optparse.OptionParser()
option_parser.add_option('-c', '--clean',
action='store_true')
option_parser.add_option('-g', '--graph',
action='store_true')
option_parser.add_option('-n', '--dry-run', '--just-print', '--recon',
action='store_true')
option_parser.add_option('-r', '--really',
action='store_true')
option_parser.add_option('-v', '--verbose',
action='count', dest='logging_level')
option_parser.set_defaults(logging_level=0)
option_parser.format_epilog = targets.format_epilog
options, args = option_parser.parse_args(argv[1:])
logging.basicConfig(format='%(asctime)s %(name)s: %(message)s',
level=logging.INFO - 10 * options.logging_level)
targets_ = []
for arg in args:
match = re.match(r'(?P<key>\w+)=(?P<value>.*)\Z', arg)
if match:
key, value = match.group('key', 'value')
if not hasattr(variables, key):
logger.error('%s is not a variable', key)
logger.debug('%s=%r', key, value)
object.__setattr__(variables, key, value)
continue
targets_.append(arg)
if not targets_:
targets_ = (targets.default.name,)
try:
for target in targets_:
target = targets.get(target)
if options.clean:
target.clean(really=options.really, recurse=True)
elif options.graph:
sys.stdout.write('digraph "%s" {\n' % (target.name,))
target.graph(sys.stdout, set())
sys.stdout.write('}\n')
else:
target.build(dry_run=options.dry_run)
except BuildError as e:
logger.error(e)
sys.exit(1)
def output(*args):
"""output captures the output of a single command. It is typically used to
set variables that only need to be set once. For example:
UNAME_A = output('uname', '-a')
If you need to capture the output of a command in a target, you should use
t.output."""
args = flatten_expand_list(args)
logger.debug(' '.join(args))
return check_output(args)
def rule(pattern):
def f(targetmaker):
rules[re.compile(pattern)] = targetmaker
return f
def target(name, *dependencies, **kwargs):
"""The @target decorator describes the action needed to build a single
target file when its dependencies are out of date. For example:
@target('hello', 'hello.c')
def hello(t):
t.run('gcc', '-o', t.name, t.dependencies)
# the above line will run gcc -o hello hello.c
See the documentation for Target to see the properties provide by the
target t."""
def f(action):
target = Target(name, action=action, dependencies=dependencies,
**kwargs)
targets.add(target)
return f
def virtual(name, *dependencies, **kwargs):
"""virtual targets are metatargets. They do not correspond to any real
file in the filesystem, even if a file with the same name already exists.
Virtual targets can be thought of as only existing for the duration of the
build. Their up-to-dateness or otherwise is independent of any existence
or up-to-dateness of any actual file in the filesystem. Typically they are
used to group actions such as "all", "build", or "test"."""
target = Target(name, dependencies=dependencies, clean=False, phony=True,
**kwargs)
targets.add(target)
def which(program):
"""Returns the full path of a given argument or `None`.
See:
http://stackoverflow.com/questions/377017/test-if-executable-exists-in-python"""
def is_exe(fpath):
return os.path.isfile(fpath) and os.access(fpath, os.X_OK)
fpath, fname = os.path.split(program)
if fpath:
if is_exe(program):
return program
else:
for path in os.environ["PATH"].split(os.pathsep):
path = path.strip('"')
exe_file = os.path.join(path, program)
if is_exe(exe_file):
return exe_file
return None

View File

@@ -0,0 +1,40 @@
var fs = require('fs');
// The number of files that we need to generate goog.require's for.
var numFiles = process.argv.length - 1;
/**
* Object used a set of found goog.provide's.
* @type {Object.<string, boolean>}
*/
var requires = {};
process.argv.forEach(function(val, index, array) {
if (index === 0) {
return;
}
fs.readFile(val, function(err, data) {
if (err) {
return;
}
var re = new RegExp('goog\\.provide\\(\'(.*)\'\\);');
data.toString().split('\n').forEach(function(line) {
var match = line.match(re);
if (match) {
requires[match[1]] = true;
}
});
if (--numFiles === 0) {
Object.keys(requires).sort().forEach(function(key) {
process.stdout.write('goog.require(\'' + key + '\');\n');
});
}
});
});

View File

@@ -16,13 +16,13 @@ Install the test dependencies (from the root of the repository):
Run the tests once with PhantomJS:
./build.py test
make test
(Note that for `npm` users, this can also be run as `npm test`.)
Run the tests in a browser:
./build.py serve
make serve
(Again for `npm` users, this is `npm start`.)
@@ -31,4 +31,4 @@ any time one of the source or spec files changes.
Tip for TDD'ers: to make PhantomJS run the test suite continuously each time
a spec file is changed you can use nosier (http://pypi.python.org/pypi/nosier)
and do `nosier -p test -p src "./build.py test"`.
and do `nosier -p test -p src "make test"`.

View File

@@ -22,7 +22,7 @@
});
</script>
<!-- This script is provided by the debug server (start with `build.py serve`) -->
<!-- This script is provided by the debug server (started with `make serve`) -->
<script type="text/javascript" src="loader.js"></script>
<script type="text/javascript">

View File

@@ -7,10 +7,10 @@ Similar to the unit tests, there are two ways to run the tests: directly in the
browser or using [SlimerJS](http://slimerjs.org/) from the command-line.
To run the tests in the browser, make sure the development server is running
(`./build.py serve`) and open the URL
(`make serve`) and open the URL
[http://localhost:3000/test_rendering/index.html](http://localhost:3000/test_rendering/index.html).
From the command-line the tests can be run with the build target `./build.py test-rendering`.
From the command-line the tests can be run with the build target `make test-rendering`.
## Adding new tests
When creating a new test case, a reference image has to be created. By appending `?generate`

View File

@@ -23,7 +23,7 @@
});
</script>
<!-- This script is provided by the debug server (start with `build.py serve`) -->
<!-- This script is provided by the debug server (started with `make serve`) -->
<script type="text/javascript" src="loader.js"></script>
<script type="text/javascript">