diff options
| author | gbrandl <devnull@localhost> | 2007-01-29 17:05:44 +0100 |
|---|---|---|
| committer | gbrandl <devnull@localhost> | 2007-01-29 17:05:44 +0100 |
| commit | 536e7dc0adab42a76dcb48ed07e3c317e4a95e10 (patch) | |
| tree | 6d78d860742f436b5953aae96a48bf2c2bcc1be0 | |
| parent | e7ab63f9c55fe71d55386779127704166e313917 (diff) | |
| download | pygments-536e7dc0adab42a76dcb48ed07e3c317e4a95e10.tar.gz | |
[svn] HTML formatter overhaul.
| -rw-r--r-- | CHANGES | 4 | ||||
| -rw-r--r-- | TODO | 3 | ||||
| -rw-r--r-- | pygments/formatters/html.py | 240 | ||||
| -rw-r--r-- | tests/test_html_formatter.py | 22 |
4 files changed, 155 insertions, 114 deletions
@@ -5,6 +5,10 @@ Version 0.7 (in development) ---------------------------- (codename to be selected, released Feb XX, 2007) +- Made the HTML formatter more flexible, and easily subclassable in order + to make it easy to implement custom wrappers, e.g. alternate line + number markup. + - Added an `outencoding` option to all formatters, making it possible to override the `encoding` (which is used by lexers and formatters) when using the command line interface. Also, if using the terminal formatter @@ -39,7 +39,4 @@ for 0.7 for 0.8 -- 1.0 -------------- -- more setuptools entrypoints (html formatter etc.) - see paste script's Commands - - add folding? would require more language-aware parsers... diff --git a/pygments/formatters/html.py b/pygments/formatters/html.py index f44053c7..368d65a2 100644 --- a/pygments/formatters/html.py +++ b/pygments/formatters/html.py @@ -5,13 +5,11 @@ Formatter for HTML output. - TODO: this is a bit of a mess and not subclassable ATM. - :copyright: 2006 by Georg Brandl, Armin Ronacher. :license: BSD, see LICENSE for more details. """ import sys, os -import cStringIO +import StringIO from pygments.formatter import Formatter from pygments.token import Token, Text, STANDARD_TYPES @@ -54,7 +52,7 @@ def _get_ttype_class(ttype): return fname + aname -DOC_TEMPLATE = '''\ +DOC_HEADER = '''\ <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd"> @@ -70,14 +68,10 @@ td.linenos { background-color: #f0f0f0; padding-right: 10px; } <body> <h2>%(title)s</h2> -%(code)s - -</body> -</html> ''' -DOC_TEMPLATE_EXTERNALCSS = '''\ +DOC_HEADER_EXTERNALCSS = '''\ <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd"> @@ -90,8 +84,9 @@ DOC_TEMPLATE_EXTERNALCSS = '''\ <body> <h2>%(title)s</h2> -%(code)s +''' +DOC_FOOTER = '''\ </body> </html> ''' @@ -288,23 +283,93 @@ class HtmlFormatter(Formatter): (arg, self.style.background_color, text_style)) return '\n'.join(lines) - def _format_nowrap(self, tokensource, outfile, lnos=False): + def _wrap_full(self, inner, outfile): + if self.cssfile: + try: + filename = outfile.name + cssfilename = os.path.join(os.path.dirname(filename), self.cssfile) + except AttributeError: + print >>sys.stderr, 'Note: Cannot determine output file name, ' \ + 'using current directory as base for the CSS file name' + cssfilename = self.cssfile + # write CSS file + try: + cf = open(cssfilename, "w") + cf.write(CSSFILE_TEMPLATE % + {'styledefs': self.get_style_defs('body')}) + cf.close() + except IOError, err: + err.strerror = 'Error writing CSS file: ' + err.strerror + raise + + yield 0, (DOC_HEADER_EXTERNALCSS % + dict(title = self.title, + cssfile = self.cssfile, + encoding = self.encoding)) + else: + yield 0, (DOC_HEADER % + dict(title = self.title, + styledefs = self.get_style_defs('body'), + encoding = self.encoding)) + + for t, line in inner: + yield t, line + yield 0, DOC_FOOTER + + def _wrap_linenos(self, inner): + dummyoutfile = StringIO.StringIO() lncount = 0 + for t, line in inner: + if t: + lncount += 1 + dummyoutfile.write(line) + + fl = self.linenostart + mw = len(str(lncount + fl - 1)) + sp = self.linenospecial + st = self.linenostep + if sp: + ls = '\n'.join([(i%st == 0 and + (i%sp == 0 and '<span class="special">%*d</span>' + or '%*d') % (mw, i) + or '') + for i in range(fl, fl + lncount)]) + else: + ls = '\n'.join([(i%st == 0 and ('%*d' % (mw, i)) or '') + for i in range(fl, fl + lncount)]) + + yield 0, ('<table><tr><td class="linenos"><pre>' + + ls + '</pre></td><td class="code">') + yield 0, dummyoutfile.getvalue() + yield 0, '</td></tr></table>' + + def _wrap_div(self, inner): + yield 0, ('<div' + (self.cssclass and ' class="%s" ' % self.cssclass) + + (self.cssstyles and ' style="%s"' % self.cssstyles) + '>') + for tup in inner: + yield tup + yield 0, '</div>\n' + + def _wrap_pre(self, inner): + yield 0, '<pre>' + for tup in inner: + yield tup + yield 0, '</pre>' + + def _format_lines(self, tokensource): + """ + Just format the tokens, without any wrapping tags. + Yield individual lines. + """ nocls = self.noclasses enc = self.encoding # for <span style=""> lookup only getcls = self.ttype2class.get c2s = self.class2style - write = outfile.write lspan = '' + line = '' for ttype, value in tokensource: - if enc: - value = value.encode(enc) - htmlvalue = escape_html(value) - if lnos: - lncount += value.count("\n") - if nocls: cclass = getcls(ttype) while cclass is None: @@ -315,97 +380,56 @@ class HtmlFormatter(Formatter): cls = self._get_css_class(ttype) cspan = cls and '<span class="%s">' % cls - if cspan == lspan: - if not cspan: - write(htmlvalue) + if enc: + value = value.encode(enc) + + parts = escape_html(value).split('\n') + + # for all but the last line + for part in parts[:-1]: + if line: + if lspan != cspan: + line += '</span>' + cspan + part + (cspan and '</span>') + '\n' + else: # both are the same + line += part + (lspan and '</span>') + '\n' + yield 1, line + line = '' else: - write(htmlvalue.replace('\n', '</span>\n' + cspan)) - elif htmlvalue: # if no value, leave old span open - if lspan: - write('</span>') - lspan = cspan - if cspan: - htmlvalue = htmlvalue.replace('\n', '</span>\n' + cspan) - write(cspan + htmlvalue) + yield 1, cspan + part + (cspan and '</span>') + '\n' + # for the last line + if line: + if lspan != cspan: + line += '</span>' + cspan + parts[-1] + lspan = cspan else: - write(htmlvalue) - if lspan: - write('</span>') - return lncount + line += parts[-1] + else: + line = cspan + parts[-1] + + if line: + yield 1, line + (lspan and '</span>') + '\n' def format(self, tokensource, outfile): - if self.nowrap: - self._format_nowrap(tokensource, outfile) - return - - realoutfile = outfile - lnos = self.linenos - full = self.full - - div = ('<div' + (self.cssclass and ' class="%s" ' % self.cssclass) - + (self.cssstyles and ' style="%s"' % self.cssstyles) + '>') - if full or lnos: - outfile = cStringIO.StringIO() - else: - outfile.write(div) - - outfile.write('<pre>') - lncount = self._format_nowrap(tokensource, outfile, lnos) - outfile.write('</pre>') - - ret = '' - if lnos: - fl = self.linenostart - mw = len(str(lncount + fl - 1)) - sp = self.linenospecial - st = self.linenostep - if sp: - ls = '\n'.join([(i%st == 0 and - (i%sp == 0 and '<span class="special">%*d</span>' - or '%*d') % (mw, i) - or '') - for i in range(fl, fl + lncount)]) - else: - ls = '\n'.join([(i%st == 0 and ('%*d' % (mw, i)) or '') - for i in range(fl, fl + lncount)]) - - ret = div + ('<table><tr>' - '<td class="linenos"><pre>' + - ls + '</pre></td><td class="code">') - ret += outfile.getvalue() - ret += '</td></tr></table>' - - if full: - if not ret: - ret = div + outfile.getvalue() + '</div>\n' - if self.cssfile: - try: - filename = realoutfile.name - cssfilename = os.path.join(os.path.dirname(filename), self.cssfile) - except AttributeError: - print >>sys.stderr, 'Note: Cannot determine output file name, ' \ - 'using current directory as base for the CSS file name' - cssfilename = self.cssfile - realoutfile.write(DOC_TEMPLATE_EXTERNALCSS % - dict(title = self.title, - cssfile = self.cssfile, - encoding = self.encoding, - code = ret)) - try: - cf = open(cssfilename, "w") - cf.write(CSSFILE_TEMPLATE % {'styledefs': - self.get_style_defs('body')}) - cf.close() - except IOError, err: - err.strerror = 'Error writing CSS file: ' + err.strerror - raise - else: - realoutfile.write(DOC_TEMPLATE % - dict(title = self.title, - styledefs = self.get_style_defs('body'), - encoding = self.encoding, - code = ret)) - elif lnos: - realoutfile.write(ret + '</div>\n') - else: - realoutfile.write('</div>\n') + """ + The formatting process uses several nested generators; which of + them are used is determined by the user's options. + + Each generator should take at least one argument, ``inner``, + and wrap the pieces of text generated by this. + + Always yield 2-tuples: (core, text). If "core" is 1, the text + is part of the original tokensource being highlighted, if it's + 0, the text is some piece of wrapping. This makes it possible to + use several different wrappers that process the original source + linewise, e.g. line number generators. + """ + source = self._format_lines(tokensource) + if not self.nowrap: + source = self._wrap_div(self._wrap_pre(source)) + if self.linenos: + source = self._wrap_linenos(source) + if self.full: + source = self._wrap_full(source, outfile) + + for t, piece in source: + outfile.write(piece) diff --git a/tests/test_html_formatter.py b/tests/test_html_formatter.py index 7e25aeaf..29d484c1 100644 --- a/tests/test_html_formatter.py +++ b/tests/test_html_formatter.py @@ -7,15 +7,31 @@ :license: BSD, see LICENSE for more details. """ +import os import unittest import StringIO import random from pygments import lexers, formatters from pygments.token import _TokenType +from pygments.formatters import HtmlFormatter +from pygments.lexers import PythonLexer class HtmlFormatterTest(unittest.TestCase): - def test_external_css(self): - # TODO: write this test. - pass +# TODO: write this test. +# def test_external_css(self): +# pass + + def test_all_options(self): + tokensource = list(PythonLexer().get_tokens(file(os.path.join(testdir, testfile)).read())) + + for optdict in [dict(nowrap=True), + dict(linenos=True), + dict(linenos=True, full=True), + dict(linenos=True, full=True, noclasses=True)]: + + outfile = StringIO.StringIO() + fmt = HtmlFormatter(**optdict) + fmt.format(tokensource, outfile) + |