2. Getting started¶
2.1. Installation¶
Transcrypt is currently tested under Windows, Linux and OSX, with Chrome, Internet Explorer and Firefox. To be able to use it, Python 3.5 has to be installed. After that, install virtualenv as explained in Jamie Matthews very clear and brief introduction. Be sure to install virtualenv for the right Python version, by using the right pip. For each Transcrypt project (or group of projects using the same Transcrypt version) create an environment as described in the referenced introduction. To install Trancrypt into that environment, activate the environment as also described there, and then type:
pip install transcrypt
from the command prompt. This is the recommended way to install Transcrypt. It is flexible and sets the proper access rights.
Alternatively, for manual installation under Windows or Linux, follow the steps below:
- Download the Transcrypt zip and unpack it anywhere you like
- Add ../Transcrypt-<version>/transcrypt to your system path
N.B. If you install Transcrypt manually, Trancrypt is started by typing run_transcrypt rather than transcrypt. This allows a pip installed Transcrypt and a manually installed Transcrypt to by used side by side selectively.
N.B.2 If you also use Numscrypt under Linux or OSX, use the MiniConda installer rather than virtualenv, as described in the Numscrypt documentation, since it will allow you to obtain the right version of NumPy.
You can test your installation as follows (replacing transcrypt by run_transcrypt if you installed manually rather than with pip):
- Go to directory ../Transcrypt-<version>/transcrypt/development/automated_tests/transcrypt
- From the command prompt run transcrypt -b autotest.py. This will compile the autotests into file autotest.js and put it into the __javascript__ subdirectory. Do NOT go to that directory (there’s no need, stay where you went at point 4)
- From the command prompt run transcrypt -r autotest.py. This will run the autotests with CPython creating file autotest.html that refers to the generated autotest.js file
- Load the autotest.html into your browser, e.g. by clicking on it (tests were done with Chrome). It will load autotest.js, run it and compare the output with what was generated by CPython. It should report no errors
To experiment with Transcrypt yourself:
- Create a directory for your experiments
- Make your own thing there, e.g. experiment1.py
- Compile with transcrypt -b experiment1.py
- Make an HTML page that will load your code in the browser. Use the HTML file generated by the autotest as an example of how to do that
- Load and run the HTML + JS
You may also want to try the demo’s.
2.1.1. Troubleshooting checklist¶
- Transcrypt was installed using pip, but import transcrypt fails. Transcrypt isn’t a library but a compiler. Install and run it as described in this chapter.
- Transcrypt reports an error containing the word ‘java’. Transcrypt produces both prettyfied and minified JavaScript output. For the minification it makes use of the Google Closure Compiler, which is included in the distribution and requires Java to run. You can check proper installation of Java by typing the word java on the command line. This should give you a list of options: Usage: java [-options] class []args...] and so on. If you can’t or won’t install Java, you can run Transcrypt without minification by using the -n command line switch.
- The static checker doesn’t find all errors it could. The static checks, performed by the PyFlakes package that’s part of the distribution, are of a ‘light’ variety. Style checks and false positives are avoided. The accent is on finding undefined identifiers and unused variables.
2.2. Your first Transcrypt program¶
Open a command prompt in the demos/hello directory and type transcrypt hello.py. Then click on hello.html to load it into your browser.
After clicking on both buttons a few times, take a look at hello.html. As you can see, eventhandlers are connected to the buttons exactly as you would do with JavaScript.
Then look at hello.py. Note that JavaScript functions like document.getElementById can be called from Python exactly as you would call them from JavaScript, but with Python data.
The minified JavaScript file including the Transcrypt runtime is only 10kB. With non-trivial programs this overhead becomes negligible. Transcrypt applications in themselves don’t need any external files. Of course you can use extensive JavaScript libraries served by content delivery networks as you normally would. But you can also make very compact stand alone applications. The most efficient thing to do is pack all functionality for a page in a single Transcrypt program. Note that the source can consist of many modules if needed and many callback entrypoints can be provided. But it is one stateful application, just like it would be on the desktop.
<script src="__javascript__/hello.js"></script>
<h2>Hello demo</h2>
<p>
<div id = "greet">...</div>
<button onclick="hello.solarSystem.greet ()">Click me repeatedly!</button>
<p>
<div id = "explain">...</div>
<button onclick="hello.solarSystem.explain ()">And click me repeatedly too!</button>
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 | from itertools import chain
class SolarSystem:
planets = [chain (planet, (index + 1,)) for index, planet in enumerate ((
('Mercury', 'hot', 2240),
('Venus', 'sulphurous', 6052),
('Earth', 'fertile', 6378),
('Mars', 'reddish', 3397),
('Jupiter', 'stormy', 71492),
('Saturn', 'ringed', 60268),
('Uranus', 'cold', 25559),
('Neptune', 'very cold', 24766)
))]
lines = (
'{} is a {} planet',
'The radius of {} is {} km',
'{} is planet nr. {} counting from the sun'
)
def __init__ (self):
self.lineIndex = 0
def greet (self):
self.planet = self.planets [int (Math.random () * len (self.planets))]
document.getElementById ('greet') .innerHTML = 'Hello {}'.format (self.planet [0])
self.explain ()
def explain (self):
document.getElementById ('explain').innerHTML = (
self.lines [self.lineIndex] .format (self.planet [0], self.planet [self.lineIndex + 1])
)
self.lineIndex = (self.lineIndex + 1) % 3
solarSystem = SolarSystem ()
|
2.3. Using sourcemaps and annotated target code¶
2.3.1. Sourcemaps¶
Sourcemaps enable debugging from the original Python source code rather then from the generated JavaScript. Transcrypt supports the use of single- and multi-level sourcemaps, using the -m switch. This means that you can source-level debug both non-minified and minified JavaScript target code. Sourcemaps are routinely tested for Google Chrome only, both under Windows and Linux, but they also have been observed to work for Firefox. Combined with the high readability of the JavaScript code generated by Transcrypt, this enables efficient debugging for real-world projects consisting of many modules.
2.3.2. Annotated target code¶
In addition to generating sourcemaps, you can use the -a switch to annotate non-minified JavaScript files with comments, referring to the original Python file names and line numbers. So even if your browser doesn’t support sourcemaps, it’s easy to find back the original Python source code location from any JavaScript statement.
// ============ Source: D:/activ_tosh/geatec/transcrypt/transcrypt/demos/hello/hello.py ============
/* 000001 */ (function () {
/* 000001 */ var chain = __init__ (__world__.itertools).chain;
/* 000003 */ var SolarSystem = __class__ ('SolarSystem', [object], {
/* 000021 */ get __init__ () {return __get__ (this, function (self) {
/* 000022 */ self.lineIndex = 0;
/* 000022 */ });},
/* 000024 */ get greet () {return __get__ (this, function (self) {
/* 000025 */ self.planet = self.planets [int (Math.random () * len (self.planets))];
/* 000026 */ document.getElementById ('greet').innerHTML = 'Hello {}'.format (self.planet [0]);
/* 000027 */ self.explain ();
/* 000027 */ });},
/* 000029 */ get explain () {return __get__ (this, function (self) {
/* 000031 */ document.getElementById ('explain').innerHTML = self.lines [self.lineIndex].format (self.planet [0], self.planet [self.lineIndex + 1]);
/* 000033 */ self.lineIndex = (self.lineIndex + 1) % 3;
/* 000033 */ });}
/* 000033 */ });
/* 000004 */ SolarSystem.planets = function () {
/* 000004 */ var __accu0__ = [];
/* 000004 */ var __iter0__ = enumerate (tuple ([tuple (['Mercury', 'hot', 2240]), tuple (['Venus', 'sulphurous', 6052]), tuple (['Earth', 'fertile', 6378]), tuple (['Mars', 'reddish', 3397]), tuple (['Jupiter', 'stormy', 71492]), tuple (['Saturn', 'ringed', 60268]), tuple (['Uranus', 'cold', 25559]), tuple (['Neptune', 'very cold', 24766])]));
/* 000004 */ for (var __index0__ = 0; __index0__ < __iter0__.length; __index0__++) {
/* 000012 */ var __left0__ = __iter0__ [__index0__];
/* 000012 */ var index = __left0__ [0];
/* 000012 */ var planet = __left0__ [1];
/* 000004 */ __accu0__.append (chain (planet, tuple ([index + 1])));
/* 000004 */ }
/* 000004 */ return __accu0__;
/* 000004 */ } ();
/* 000015 */ SolarSystem.lines = tuple (['{} is a {} planet', 'The radius of {} is {} km', '{} is planet nr. {} counting from the sun']);
/* 000035 */ var solarSystem = SolarSystem ();
/* 000035 */ __pragma__ ('<use>' +
/* 000035 */ 'itertools' +
/* 000035 */ '</use>')
/* 000035 */ __pragma__ ('<all>')
/* 000035 */ __all__.SolarSystem = SolarSystem;
/* 000035 */ __all__.solarSystem = solarSystem;
/* 000035 */ __pragma__ ('</all>')
/* 000035 */ }) ();
/* 000035 */ return __all__;
Source code annotation only happens for Python sources, not for JavaScript-only modules, that have a trivial correspondence between non-minified target code and source code.
2.4. Compiling for node.js¶
Transcrypt will allow you to target node.js while writing your server code in Python. This opens up the exploding world of node.js libraries from your favorite programming language. In the demo/nodejs_demo subdirectory of the installation, you’ll find the following trivial example of a node.js server app:
# Compile with p. command line switch (see docs).
# The example will be served at URL: http://localhost:8080 in your browser
http = require ('http')
class Demo:
texts = (
'Welcome to the world of node.js',
'You can have your cake and eat it',
'Use node\'s ecosystem while programming in Python',
'Using node.js from Transcrypt is easy',
'Take a Python ride into the node.js world'
)
def __init__ (self, port):
print ('Demo server started on port', port)
self.server = http.createServer (self.serve)
self.server.listen (port)
self.oldIndex = 0
self.newIndex = 0
self.count = 0
def serve (self, request, response):
response.writeHead (200)
print ('Serving page', self.count)
self.count += 1
while self.newIndex == self.oldIndex:
self.newIndex = int (Math.random () * len (self.texts))
self.oldIndex = self.newIndex
response.end ('<h1>{}</h1>'.format (
self.texts [self.newIndex]
))
demo = Demo (8080)
Follow the steps below to run this demo:
- Install node.js from https://nodejs.org
- Open a node.js command prompt
- Go to the demo/nodejs_demo directory
- Compile the demo with transcrypt -b -p .none nodejs_demo.py, to generate an orphan module rather than a child of window
- Go to demo/nodejs_demo/__javascript__ directory
- Type node nodejs_demo.js (or node nodejs_demo.min.js if you want to run the minified version)
- In your browser, view the result at http://localhost:8080
- Repeatedly reload the page to see the text change (Google Chrome may actually reload twice)
2.5. Available command line switches¶
The available command line switches will be shown if you run transcript -h. They are specified in the source code of Transcrypt as follows:
class CommandArgs:
def parse (self):
self.argParser = argparse.ArgumentParser ()
self.argParser.add_argument ('source', nargs='?', help = ".py file containing source code of main module")
self.argParser.add_argument ('-a', '--anno', help = "annotate target files that were compiled from Python with source file names and source line numbers", action = 'store_true')
self.argParser.add_argument ('-b', '--build', help = "rebuild all target files from scratch", action = 'store_true')
self.argParser.add_argument ('-c', '--check', help = "perform static check as part of compilation", action = 'store_true')
self.argParser.add_argument ('-d', '--dmap', help = "dump human readable source map", action = 'store_true')
self.argParser.add_argument ('-e', '--extex', help = "extended exception reports", action = 'store_true')
self.argParser.add_argument ('-f', '--fcall', help = "enable fastcall mechanism by default. You can also use __pragma__ ('fcal') and __pragma__ (\'nofcall\')", action = 'store_true')
self.argParser.add_argument ('-i', '--iconv', help = "enable automatic conversion to iterable by default. DISADVISED, since it will result in a type check for each for-loop. Preferable use __pragma__ ('iconv') and __pragma__ (\'noiconv\') to enable automatic conversion locally", action = 'store_true')
self.argParser.add_argument ('-j', '--jskeys', help = "interpret {key: 'value'} as {'key': 'value'} and forbid {key (): 'value'}, as JavaScript does. DISADVISED, since it's less flexible than the Python interpretation. Use {'key': 'value'} explicitly if you want literal keys", action = 'store_true')
self.argParser.add_argument ('-k', '--kwargs', help = "enable keyword arguments by default. In general this is DISADVISED, use __pragma__ ('kwargs') and __pragma__('nokwargs') locally instead to prevent bloated code", action = 'store_true')
self.argParser.add_argument ('-l', '--license', help = "show license", action = 'store_true')
self.argParser.add_argument ('-m', '--map', help = "generate source map", action = 'store_true')
self.argParser.add_argument ('-n', '--nomin', help = "no minification", action = 'store_true')
self.argParser.add_argument ('-o', '--opov', help = "enable operator overloading by default. In general this is DISADVISED, use __pragma__ ('opov') and __pragma__('noopov') locally instead to prevent slow code", action = 'store_true')
self.argParser.add_argument ('-p', '--parent', nargs='?', help = "object that will hold module, default is window. Use -p .none to generate orphan module, e.g. for use in node.js. Use -p .user to generate module that has to be explicitly initialized by calling <modulename> (), e.g. after the full page has loaded")
self.argParser.add_argument ('-r', '--run', help = "run source file rather than compiling it", action = 'store_true')
self.argParser.add_argument ('-t', '--tree', help = "dump syntax tree", action = 'store_true')
self.argParser.add_argument ('-v', '--verbose', help = "show all messages", action = 'store_true')
self.__dict__.update (self.argParser.parse_args () .__dict__)
if not (self.license or self.source):
log (True, self.argParser.format_usage () .capitalize ())
sys.exit (1)
global extraLines
extraLines = [
# Make identifier __pragma__ known to static checker
# It was only known in JavaScript from __core__.mod.js, which the checker doesn't see
# __pragma__ ('<all>') in JavaScript requires it to remain a function, as it was in the core
# It can't be skipped, since it has to precede __pragma__ ('skip'), to make the checker accept that
'def __pragma__ (): pass',
# Make __include__ known to the static checker
'__pragma__ (\'skip\')',
'__new__ = __include__ = 0',
'__pragma__ (\'noskip\')',
''
] if commandArgs.check else []
global nrOfExtraLines
nrOfExtraLines = max (len (extraLines) - 1, 0) # Last line only serves to force linefeed
extraLines = '\n'.join (extraLines)
If static checking is enabled, insert dummy definitions of global JavaScript variables between __pragma__ (‘skip’) and __pragma__ (‘noskip’) to prevent needless complaints of the checker. The static checks are geared towards avoiding false alarms, and mainly check undefined names and unused variables. Style checks are deliberately avoided.