Changeset - e46d0a90556e
[Not reviewed]
beta
0 1 0
Marcin Kuzminski - 13 years ago 2012-10-27 14:33:29
marcin@python-works.com
fixes #612 Double quotes to Single quotes result in 4; to 9; in a visual Diff. Markupsafe.escape method usage
was causing some unpredictable behavior
1 file changed with 6 insertions and 2 deletions:
0 comments (0 inline, 0 general)
rhodecode/lib/diffs.py
Show inline comments
 
# -*- coding: utf-8 -*-
 
"""
 
    rhodecode.lib.diffs
 
    ~~~~~~~~~~~~~~~~~~~
 

	
 
    Set of diffing helpers, previously part of vcs
 

	
 

	
 
    :created_on: Dec 4, 2011
 
    :author: marcink
 
    :copyright: (C) 2010-2012 Marcin Kuzminski <marcin@python-works.com>
 
    :original copyright: 2007-2008 by Armin Ronacher
 
    :license: GPLv3, see COPYING for more details.
 
"""
 
# This program is free software: you can redistribute it and/or modify
 
# it under the terms of the GNU General Public License as published by
 
# the Free Software Foundation, either version 3 of the License, or
 
# (at your option) any later version.
 
#
 
# This program is distributed in the hope that it will be useful,
 
# but WITHOUT ANY WARRANTY; without even the implied warranty of
 
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 
# GNU General Public License for more details.
 
#
 
# You should have received a copy of the GNU General Public License
 
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
 

	
 
import re
 
import difflib
 
import markupsafe
 
import logging
 

	
 
from itertools import tee, imap
 

	
 
from mercurial import patch
 
from mercurial.mdiff import diffopts
 
from mercurial.bundlerepo import bundlerepository
 

	
 
from pylons.i18n.translation import _
 

	
 
from rhodecode.lib.compat import BytesIO
 
from rhodecode.lib.vcs.utils.hgcompat import localrepo
 
from rhodecode.lib.vcs.exceptions import VCSError
 
from rhodecode.lib.vcs.nodes import FileNode, SubModuleNode
 
from rhodecode.lib.vcs.backends.base import EmptyChangeset
 
from rhodecode.lib.helpers import escape
 
from rhodecode.lib.utils import make_ui
 
from rhodecode.lib.utils2 import safe_unicode
 

	
 
log = logging.getLogger(__name__)
 

	
 

	
 
def wrap_to_table(str_):
 
    return '''<table class="code-difftable">
 
                <tr class="line no-comment">
 
                <td class="lineno new"></td>
 
                <td class="code no-comment"><pre>%s</pre></td>
 
                </tr>
 
              </table>''' % str_
 

	
 

	
 
def wrapped_diff(filenode_old, filenode_new, cut_off_limit=None,
 
                ignore_whitespace=True, line_context=3,
 
                enable_comments=False):
 
    """
 
    returns a wrapped diff into a table, checks for cut_off_limit and presents
 
    proper message
 
    """
 

	
 
    if filenode_old is None:
 
        filenode_old = FileNode(filenode_new.path, '', EmptyChangeset())
 

	
 
    if filenode_old.is_binary or filenode_new.is_binary:
 
        diff = wrap_to_table(_('binary file'))
 
        stats = (0, 0)
 
        size = 0
 

	
 
    elif cut_off_limit != -1 and (cut_off_limit is None or
 
@@ -126,97 +125,99 @@ def get_gitdiff(filenode_old, filenode_n
 
    repo = filenode_new.changeset.repository
 
    old_raw_id = getattr(filenode_old.changeset, 'raw_id', repo.EMPTY_CHANGESET)
 
    new_raw_id = getattr(filenode_new.changeset, 'raw_id', repo.EMPTY_CHANGESET)
 

	
 
    vcs_gitdiff = repo.get_diff(old_raw_id, new_raw_id, filenode_new.path,
 
                                 ignore_whitespace, context)
 
    return vcs_gitdiff
 

	
 

	
 
class DiffProcessor(object):
 
    """
 
    Give it a unified diff and it returns a list of the files that were
 
    mentioned in the diff together with a dict of meta information that
 
    can be used to render it in a HTML template.
 
    """
 
    _chunk_re = re.compile(r'@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@(.*)')
 
    _newline_marker = '\\ No newline at end of file\n'
 

	
 
    def __init__(self, diff, differ='diff', format='gitdiff'):
 
        """
 
        :param diff:   a text in diff format or generator
 
        :param format: format of diff passed, `udiff` or `gitdiff`
 
        """
 
        if isinstance(diff, basestring):
 
            diff = [diff]
 

	
 
        self.__udiff = diff
 
        self.__format = format
 
        self.adds = 0
 
        self.removes = 0
 

	
 
        if isinstance(self.__udiff, basestring):
 
            self.lines = iter(self.__udiff.splitlines(1))
 

	
 
        elif self.__format == 'gitdiff':
 
            udiff_copy = self.copy_iterator()
 
            self.lines = imap(self.escaper, self._parse_gitdiff(udiff_copy))
 
        else:
 
            udiff_copy = self.copy_iterator()
 
            self.lines = imap(self.escaper, udiff_copy)
 

	
 
        # Select a differ.
 
        if differ == 'difflib':
 
            self.differ = self._highlight_line_difflib
 
        else:
 
            self.differ = self._highlight_line_udiff
 

	
 
    def escaper(self, string):
 
        return markupsafe.escape(string)
 
        return string.replace('&', '&amp;')\
 
                .replace('<', '&lt;')\
 
                .replace('>', '&gt;')
 

	
 
    def copy_iterator(self):
 
        """
 
        make a fresh copy of generator, we should not iterate thru
 
        an original as it's needed for repeating operations on
 
        this instance of DiffProcessor
 
        """
 
        self.__udiff, iterator_copy = tee(self.__udiff)
 
        return iterator_copy
 

	
 
    def _extract_rev(self, line1, line2):
 
        """
 
        Extract the operation (A/M/D), filename and revision hint from a line.
 
        """
 

	
 
        try:
 
            if line1.startswith('--- ') and line2.startswith('+++ '):
 
                l1 = line1[4:].split(None, 1)
 
                old_filename = (l1[0].replace('a/', '', 1)
 
                                if len(l1) >= 1 else None)
 
                old_rev = l1[1] if len(l1) == 2 else 'old'
 

	
 
                l2 = line2[4:].split(None, 1)
 
                new_filename = (l2[0].replace('b/', '', 1)
 
                                if len(l1) >= 1 else None)
 
                new_rev = l2[1] if len(l2) == 2 else 'new'
 

	
 
                filename = (old_filename
 
                            if old_filename != '/dev/null' else new_filename)
 

	
 
                operation = 'D' if new_filename == '/dev/null' else None
 
                if not operation:
 
                    operation = 'M' if old_filename != '/dev/null' else 'A'
 

	
 
                return operation, filename, new_rev, old_rev
 
        except (ValueError, IndexError):
 
            pass
 

	
 
        return None, None, None, None
 

	
 
    def _parse_gitdiff(self, diffiterator):
 
        def line_decoder(l):
 
            if l.startswith('+') and not l.startswith('+++'):
 
                self.adds += 1
 
            elif l.startswith('-') and not l.startswith('---'):
 
                self.removes += 1
 
            return safe_unicode(l)
 

	
 
@@ -255,119 +256,121 @@ class DiffProcessor(object):
 
            oldfrag = ''.join(oldwords[i1:i2])
 
            newfrag = ''.join(newwords[j1:j2])
 
            if tag != 'equal':
 
                if oldfrag:
 
                    oldfrag = '<del>%s</del>' % oldfrag
 
                if newfrag:
 
                    newfrag = '<ins>%s</ins>' % newfrag
 
            oldfragments.append(oldfrag)
 
            newfragments.append(newfrag)
 

	
 
        old['line'] = "".join(oldfragments)
 
        new['line'] = "".join(newfragments)
 

	
 
    def _highlight_line_udiff(self, line, next_):
 
        """
 
        Highlight inline changes in both lines.
 
        """
 
        start = 0
 
        limit = min(len(line['line']), len(next_['line']))
 
        while start < limit and line['line'][start] == next_['line'][start]:
 
            start += 1
 
        end = -1
 
        limit -= start
 
        while -end <= limit and line['line'][end] == next_['line'][end]:
 
            end -= 1
 
        end += 1
 
        if start or end:
 
            def do(l):
 
                last = end + len(l['line'])
 
                if l['action'] == 'add':
 
                    tag = 'ins'
 
                else:
 
                    tag = 'del'
 
                l['line'] = '%s<%s>%s</%s>%s' % (
 
                    l['line'][:start],
 
                    tag,
 
                    l['line'][start:last],
 
                    tag,
 
                    l['line'][last:]
 
                )
 
            do(line)
 
            do(next_)
 

	
 
    def _parse_udiff(self, inline_diff=True):
 
        """
 
        Parse the diff an return data for the template.
 
        """
 
        lineiter = self.lines
 

	
 
        files = []
 
        try:
 
            line = lineiter.next()
 
            while 1:
 
                # continue until we found the old file
 
                if not line.startswith('--- '):
 
                    line = lineiter.next()
 
                    continue
 

	
 
                chunks = []
 
                stats = [0, 0]
 
                operation, filename, old_rev, new_rev = \
 
                    self._extract_rev(line, lineiter.next())
 
                files.append({
 
                    'filename':         filename,
 
                    'old_revision':     old_rev,
 
                    'new_revision':     new_rev,
 
                    'chunks':           chunks,
 
                    'operation':        operation,
 
                    'stats':            stats,
 
                })
 

	
 
                line = lineiter.next()
 

	
 
                while line:
 
                    match = self._chunk_re.match(line)
 
                    if not match:
 
                        break
 

	
 
                    lines = []
 
                    chunks.append(lines)
 

	
 
                    old_line, old_end, new_line, new_end = \
 
                        [int(x or 1) for x in match.groups()[:-1]]
 
                    old_line -= 1
 
                    new_line -= 1
 
                    gr = match.groups()
 
                    context = len(gr) == 5
 
                    old_end += old_line
 
                    new_end += new_line
 

	
 
                    if context:
 
                        # skip context only if it's first line
 
                        if int(gr[0]) > 1:
 
                            lines.append({
 
                                'old_lineno': '...',
 
                                'new_lineno': '...',
 
                                'action':     'context',
 
                                'line':       line,
 
                            })
 

	
 
                    line = lineiter.next()
 

	
 
                    while old_line < old_end or new_line < new_end:
 
                        if line:
 
                            command = line[0]
 
                            if command in ['+', '-', ' ']:
 
                                #only modify the line if it's actually a diff
 
                                # thing
 
                                line = line[1:]
 
                        else:
 
                            command = ' '
 

	
 
                        affects_old = affects_new = False
 

	
 
                        # ignore those if we don't expect them
 
                        if command in '#@':
 
                            continue
 
                        elif command == '+':
 
                            affects_new = True
 
                            action = 'add'
 
                            stats[0] += 1
 
@@ -506,96 +509,97 @@ class DiffProcessor(object):
 
                        'oldline_no': change['new_lineno']
 
                    }
 
                    cond_old = (change['old_lineno'] != '...' and
 
                                change['old_lineno'])
 
                    cond_new = (change['new_lineno'] != '...' and
 
                                change['new_lineno'])
 
                    if cond_old:
 
                        anchor_old_id = 'id="%s"' % anchor_old
 
                    if cond_new:
 
                        anchor_new_id = 'id="%s"' % anchor_new
 
                    ###########################################################
 
                    # OLD LINE NUMBER
 
                    ###########################################################
 
                    _html.append('''\t<td %(a_id)s class="%(olc)s">''' % {
 
                        'a_id': anchor_old_id,
 
                        'olc': old_lineno_class
 
                    })
 

	
 
                    _html.append('''%(link)s''' % {
 
                        'link': _link_to_if(True, change['old_lineno'],
 
                                            '#%s' % anchor_old)
 
                    })
 
                    _html.append('''</td>\n''')
 
                    ###########################################################
 
                    # NEW LINE NUMBER
 
                    ###########################################################
 

	
 
                    _html.append('''\t<td %(a_id)s class="%(nlc)s">''' % {
 
                        'a_id': anchor_new_id,
 
                        'nlc': new_lineno_class
 
                    })
 

	
 
                    _html.append('''%(link)s''' % {
 
                        'link': _link_to_if(True, change['new_lineno'],
 
                                            '#%s' % anchor_new)
 
                    })
 
                    _html.append('''</td>\n''')
 
                    ###########################################################
 
                    # CODE
 
                    ###########################################################
 
                    comments = '' if enable_comments else 'no-comment'
 
                    _html.append('''\t<td class="%(cc)s %(inc)s">''' % {
 
                        'cc': code_class,
 
                        'inc': comments
 
                    })
 
                    _html.append('''\n\t\t<pre>%(code)s</pre>\n''' % {
 
                        'code': change['line']
 
                    })
 

	
 
                    _html.append('''\t</td>''')
 
                    _html.append('''\n</tr>\n''')
 
        _html.append('''</table>''')
 
        if _html_empty:
 
            return None
 
        return ''.join(_html)
 

	
 
    def stat(self):
 
        """
 
        Returns tuple of added, and removed lines for this instance
 
        """
 
        return self.adds, self.removes
 

	
 

	
 
class InMemoryBundleRepo(bundlerepository):
 
    def __init__(self, ui, path, bundlestream):
 
        self._tempparent = None
 
        localrepo.localrepository.__init__(self, ui, path)
 
        self.ui.setconfig('phases', 'publish', False)
 

	
 
        self.bundle = bundlestream
 

	
 
        # dict with the mapping 'filename' -> position in the bundle
 
        self.bundlefilespos = {}
 

	
 

	
 
def differ(org_repo, org_ref, other_repo, other_ref, discovery_data=None,
 
           bundle_compare=False):
 
    """
 
    General differ between branches, bookmarks or separate but releated
 
    repositories
 

	
 
    :param org_repo:
 
    :type org_repo:
 
    :param org_ref:
 
    :type org_ref:
 
    :param other_repo:
 
    :type other_repo:
 
    :param other_ref:
 
    :type other_ref:
 
    """
 

	
 
    bundlerepo = None
 
    ignore_whitespace = False
 
    context = 3
 
    org_repo = org_repo.scm_instance._repo
 
    other_repo = other_repo.scm_instance._repo
 
    opts = diffopts(git=True, ignorews=ignore_whitespace, context=context)
0 comments (0 inline, 0 general)