mdiff: add a hunkinrange helper function

This factors out hunk filtering logic by line range that is similar in
mdiff.blocksinrange() and hgweb.webutil.diffs().

mercurial/hgweb/webutil.py

@@ -1,628 +1,628 b''
 # hgweb/webutil.py - utility library for the web interface.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import copy
 import difflib
 import os
 import re
 
 from ..i18n import _
 from ..node import hex, nullid, short
 
 from .common import (
     ErrorResponse,
     HTTP_BAD_REQUEST,
     HTTP_NOT_FOUND,
     paritygen,
 )
 
 from .. import (
     context,
     error,
     match,
+    mdiff,
     patch,
     pathutil,
     templatefilters,
     ui as uimod,
     util,
 )
 
 def up(p):
     if p[0] != "/":
         p = "/" + p
     if p[-1] == "/":
         p = p[:-1]
     up = os.path.dirname(p)
     if up == "/":
         return "/"
     return up + "/"
 
 def _navseq(step, firststep=None):
     if firststep:
         yield firststep
         if firststep >= 20 and firststep <= 40:
             firststep = 50
             yield firststep
         assert step > 0
         assert firststep > 0
         while step <= firststep:
             step *= 10
     while True:
         yield 1 * step
         yield 3 * step
         step *= 10
 
 class revnav(object):
 
     def __init__(self, repo):
         """Navigation generation object
 
         :repo: repo object we generate nav for
         """
         # used for hex generation
         self._revlog = repo.changelog
 
     def __nonzero__(self):
         """return True if any revision to navigate over"""
         return self._first() is not None
 
     __bool__ = __nonzero__
 
     def _first(self):
         """return the minimum non-filtered changeset or None"""
         try:
             return next(iter(self._revlog))
         except StopIteration:
             return None
 
     def hex(self, rev):
         return hex(self._revlog.node(rev))
 
     def gen(self, pos, pagelen, limit):
         """computes label and revision id for navigation link
 
         :pos: is the revision relative to which we generate navigation.
         :pagelen: the size of each navigation page
         :limit: how far shall we link
 
         The return is:
             - a single element tuple
             - containing a dictionary with a `before` and `after` key
             - values are generator functions taking arbitrary number of kwargs
             - yield items are dictionaries with `label` and `node` keys
         """
         if not self:
             # empty repo
             return ({'before': (), 'after': ()},)
 
         targets = []
         for f in _navseq(1, pagelen):
             if f > limit:
                 break
             targets.append(pos + f)
             targets.append(pos - f)
         targets.sort()
 
         first = self._first()
         navbefore = [("(%i)" % first, self.hex(first))]
         navafter = []
         for rev in targets:
             if rev not in self._revlog:
                 continue
             if pos < rev < limit:
                 navafter.append(("+%d" % abs(rev - pos), self.hex(rev)))
             if 0 < rev < pos:
                 navbefore.append(("-%d" % abs(rev - pos), self.hex(rev)))
 
 
         navafter.append(("tip", "tip"))
 
         data = lambda i: {"label": i[0], "node": i[1]}
         return ({'before': lambda **map: (data(i) for i in navbefore),
                  'after':  lambda **map: (data(i) for i in navafter)},)
 
 class filerevnav(revnav):
 
     def __init__(self, repo, path):
         """Navigation generation object
 
         :repo: repo object we generate nav for
         :path: path of the file we generate nav for
         """
         # used for iteration
         self._changelog = repo.unfiltered().changelog
         # used for hex generation
         self._revlog = repo.file(path)
 
     def hex(self, rev):
         return hex(self._changelog.node(self._revlog.linkrev(rev)))
 
 class _siblings(object):
     def __init__(self, siblings=None, hiderev=None):
         if siblings is None:
             siblings = []
         self.siblings = [s for s in siblings if s.node() != nullid]
         if len(self.siblings) == 1 and self.siblings[0].rev() == hiderev:
             self.siblings = []
 
     def __iter__(self):
         for s in self.siblings:
             d = {
                 'node': s.hex(),
                 'rev': s.rev(),
                 'user': s.user(),
                 'date': s.date(),
                 'description': s.description(),
                 'branch': s.branch(),
             }
             if util.safehasattr(s, 'path'):
                 d['file'] = s.path()
             yield d
 
     def __len__(self):
         return len(self.siblings)
 
 def annotate(fctx, ui):
     diffopts = patch.difffeatureopts(ui, untrusted=True,
                                      section='annotate', whitespace=True)
     return fctx.annotate(follow=True, linenumber=True, diffopts=diffopts)
 
 def parents(ctx, hide=None):
     if isinstance(ctx, context.basefilectx):
         introrev = ctx.introrev()
         if ctx.changectx().rev() != introrev:
             return _siblings([ctx.repo()[introrev]], hide)
     return _siblings(ctx.parents(), hide)
 
 def children(ctx, hide=None):
     return _siblings(ctx.children(), hide)
 
 def renamelink(fctx):
     r = fctx.renamed()
     if r:
         return [{'file': r[0], 'node': hex(r[1])}]
     return []
 
 def nodetagsdict(repo, node):
     return [{"name": i} for i in repo.nodetags(node)]
 
 def nodebookmarksdict(repo, node):
     return [{"name": i} for i in repo.nodebookmarks(node)]
 
 def nodebranchdict(repo, ctx):
     branches = []
     branch = ctx.branch()
     # If this is an empty repo, ctx.node() == nullid,
     # ctx.branch() == 'default'.
     try:
         branchnode = repo.branchtip(branch)
     except error.RepoLookupError:
         branchnode = None
     if branchnode == ctx.node():
         branches.append({"name": branch})
     return branches
 
 def nodeinbranch(repo, ctx):
     branches = []
     branch = ctx.branch()
     try:
         branchnode = repo.branchtip(branch)
     except error.RepoLookupError:
         branchnode = None
     if branch != 'default' and branchnode != ctx.node():
         branches.append({"name": branch})
     return branches
 
 def nodebranchnodefault(ctx):
     branches = []
     branch = ctx.branch()
     if branch != 'default':
         branches.append({"name": branch})
     return branches
 
 def showtag(repo, tmpl, t1, node=nullid, **args):
     for t in repo.nodetags(node):
         yield tmpl(t1, tag=t, **args)
 
 def showbookmark(repo, tmpl, t1, node=nullid, **args):
     for t in repo.nodebookmarks(node):
         yield tmpl(t1, bookmark=t, **args)
 
 def branchentries(repo, stripecount, limit=0):
     tips = []
     heads = repo.heads()
     parity = paritygen(stripecount)
     sortkey = lambda item: (not item[1], item[0].rev())
 
     def entries(**map):
         count = 0
         if not tips:
             for tag, hs, tip, closed in repo.branchmap().iterbranches():
                 tips.append((repo[tip], closed))
         for ctx, closed in sorted(tips, key=sortkey, reverse=True):
             if limit > 0 and count >= limit:
                 return
             count += 1
             if closed:
                 status = 'closed'
             elif ctx.node() not in heads:
                 status = 'inactive'
             else:
                 status = 'open'
             yield {
                 'parity': next(parity),
                 'branch': ctx.branch(),
                 'status': status,
                 'node': ctx.hex(),
                 'date': ctx.date()
             }
 
     return entries
 
 def cleanpath(repo, path):
     path = path.lstrip('/')
     return pathutil.canonpath(repo.root, '', path)
 
 def changeidctx(repo, changeid):
     try:
         ctx = repo[changeid]
     except error.RepoError:
         man = repo.manifestlog._revlog
         ctx = repo[man.linkrev(man.rev(man.lookup(changeid)))]
 
     return ctx
 
 def changectx(repo, req):
     changeid = "tip"
     if 'node' in req.form:
         changeid = req.form['node'][0]
         ipos = changeid.find(':')
         if ipos != -1:
             changeid = changeid[(ipos + 1):]
     elif 'manifest' in req.form:
         changeid = req.form['manifest'][0]
 
     return changeidctx(repo, changeid)
 
 def basechangectx(repo, req):
     if 'node' in req.form:
         changeid = req.form['node'][0]
         ipos = changeid.find(':')
         if ipos != -1:
             changeid = changeid[:ipos]
             return changeidctx(repo, changeid)
 
     return None
 
 def filectx(repo, req):
     if 'file' not in req.form:
         raise ErrorResponse(HTTP_NOT_FOUND, 'file not given')
     path = cleanpath(repo, req.form['file'][0])
     if 'node' in req.form:
         changeid = req.form['node'][0]
     elif 'filenode' in req.form:
         changeid = req.form['filenode'][0]
     else:
         raise ErrorResponse(HTTP_NOT_FOUND, 'node or filenode not given')
     try:
         fctx = repo[changeid][path]
     except error.RepoError:
         fctx = repo.filectx(path, fileid=changeid)
 
     return fctx
 
 def linerange(req):
     linerange = req.form.get('linerange')
     if linerange is None:
         return None
     if len(linerange) > 1:
         raise ErrorResponse(HTTP_BAD_REQUEST,
                             'redundant linerange parameter')
     try:
         fromline, toline = map(int, linerange[0].split(':', 1))
     except ValueError:
         raise ErrorResponse(HTTP_BAD_REQUEST,
                             'invalid linerange parameter')
     try:
         return util.processlinerange(fromline, toline)
     except error.ParseError as exc:
         raise ErrorResponse(HTTP_BAD_REQUEST, str(exc))
 
 def formatlinerange(fromline, toline):
     return '%d:%d' % (fromline + 1, toline)
 
 def commonentry(repo, ctx):
     node = ctx.node()
     return {
         'rev': ctx.rev(),
         'node': hex(node),
         'author': ctx.user(),
         'desc': ctx.description(),
         'date': ctx.date(),
         'extra': ctx.extra(),
         'phase': ctx.phasestr(),
         'branch': nodebranchnodefault(ctx),
         'inbranch': nodeinbranch(repo, ctx),
         'branches': nodebranchdict(repo, ctx),
         'tags': nodetagsdict(repo, node),
         'bookmarks': nodebookmarksdict(repo, node),
         'parent': lambda **x: parents(ctx),
         'child': lambda **x: children(ctx),
     }
 
 def changelistentry(web, ctx, tmpl):
     '''Obtain a dictionary to be used for entries in a changelist.
 
     This function is called when producing items for the "entries" list passed
     to the "shortlog" and "changelog" templates.
     '''
     repo = web.repo
     rev = ctx.rev()
     n = ctx.node()
     showtags = showtag(repo, tmpl, 'changelogtag', n)
     files = listfilediffs(tmpl, ctx.files(), n, web.maxfiles)
 
     entry = commonentry(repo, ctx)
     entry.update(
         allparents=lambda **x: parents(ctx),
         parent=lambda **x: parents(ctx, rev - 1),
         child=lambda **x: children(ctx, rev + 1),
         changelogtag=showtags,
         files=files,
     )
     return entry
 
 def symrevorshortnode(req, ctx):
     if 'node' in req.form:
         return templatefilters.revescape(req.form['node'][0])
     else:
         return short(ctx.node())
 
 def changesetentry(web, req, tmpl, ctx):
     '''Obtain a dictionary to be used to render the "changeset" template.'''
 
     showtags = showtag(web.repo, tmpl, 'changesettag', ctx.node())
     showbookmarks = showbookmark(web.repo, tmpl, 'changesetbookmark',
                                  ctx.node())
     showbranch = nodebranchnodefault(ctx)
 
     files = []
     parity = paritygen(web.stripecount)
     for blockno, f in enumerate(ctx.files()):
         template = f in ctx and 'filenodelink' or 'filenolink'
         files.append(tmpl(template,
                           node=ctx.hex(), file=f, blockno=blockno + 1,
                           parity=next(parity)))
 
     basectx = basechangectx(web.repo, req)
     if basectx is None:
         basectx = ctx.p1()
 
     style = web.config('web', 'style', 'paper')
     if 'style' in req.form:
         style = req.form['style'][0]
 
     diff = diffs(web, tmpl, ctx, basectx, None, style)
 
     parity = paritygen(web.stripecount)
     diffstatsgen = diffstatgen(ctx, basectx)
     diffstats = diffstat(tmpl, ctx, diffstatsgen, parity)
 
     return dict(
         diff=diff,
         symrev=symrevorshortnode(req, ctx),
         basenode=basectx.hex(),
         changesettag=showtags,
         changesetbookmark=showbookmarks,
         changesetbranch=showbranch,
         files=files,
         diffsummary=lambda **x: diffsummary(diffstatsgen),
         diffstat=diffstats,
         archives=web.archivelist(ctx.hex()),
         **commonentry(web.repo, ctx))
 
 def listfilediffs(tmpl, files, node, max):
     for f in files[:max]:
         yield tmpl('filedifflink', node=hex(node), file=f)
     if len(files) > max:
         yield tmpl('fileellipses')
 
 def diffs(web, tmpl, ctx, basectx, files, style, linerange=None,
           lineidprefix=''):
 
     def prettyprintlines(lines, blockno):
         for lineno, l in enumerate(lines, 1):
             difflineno = "%d.%d" % (blockno, lineno)
             if l.startswith('+'):
                 ltype = "difflineplus"
             elif l.startswith('-'):
                 ltype = "difflineminus"
             elif l.startswith('@'):
                 ltype = "difflineat"
             else:
                 ltype = "diffline"
             yield tmpl(ltype,
                        line=l,
                        lineno=lineno,
                        lineid=lineidprefix + "l%s" % difflineno,
                        linenumber="% 8s" % difflineno)
 
     repo = web.repo
     if files:
         m = match.exact(repo.root, repo.getcwd(), files)
     else:
         m = match.always(repo.root, repo.getcwd())
 
     diffopts = patch.diffopts(repo.ui, untrusted=True)
     node1 = basectx.node()
     node2 = ctx.node()
     parity = paritygen(web.stripecount)
 
     diffhunks = patch.diffhunks(repo, node1, node2, m, opts=diffopts)
     for blockno, (header, hunks) in enumerate(diffhunks, 1):
         if style != 'raw':
             header = header[1:]
         lines = [h + '\n' for h in header]
         for hunkrange, hunklines in hunks:
             if linerange is not None and hunkrange is not None:
                 s1, l1, s2, l2 = hunkrange
-                lb, ub = linerange
-                if not (lb < s2 + l2 and ub > s2):
+                if not mdiff.hunkinrange((s2, l2), linerange):
                     continue
             lines.extend(hunklines)
         if lines:
             yield tmpl('diffblock', parity=next(parity), blockno=blockno,
                        lines=prettyprintlines(lines, blockno))
 
 def compare(tmpl, context, leftlines, rightlines):
     '''Generator function that provides side-by-side comparison data.'''
 
     def compline(type, leftlineno, leftline, rightlineno, rightline):
         lineid = leftlineno and ("l%s" % leftlineno) or ''
         lineid += rightlineno and ("r%s" % rightlineno) or ''
         return tmpl('comparisonline',
                     type=type,
                     lineid=lineid,
                     leftlineno=leftlineno,
                     leftlinenumber="% 6s" % (leftlineno or ''),
                     leftline=leftline or '',
                     rightlineno=rightlineno,
                     rightlinenumber="% 6s" % (rightlineno or ''),
                     rightline=rightline or '')
 
     def getblock(opcodes):
         for type, llo, lhi, rlo, rhi in opcodes:
             len1 = lhi - llo
             len2 = rhi - rlo
             count = min(len1, len2)
             for i in xrange(count):
                 yield compline(type=type,
                                leftlineno=llo + i + 1,
                                leftline=leftlines[llo + i],
                                rightlineno=rlo + i + 1,
                                rightline=rightlines[rlo + i])
             if len1 > len2:
                 for i in xrange(llo + count, lhi):
                     yield compline(type=type,
                                    leftlineno=i + 1,
                                    leftline=leftlines[i],
                                    rightlineno=None,
                                    rightline=None)
             elif len2 > len1:
                 for i in xrange(rlo + count, rhi):
                     yield compline(type=type,
                                    leftlineno=None,
                                    leftline=None,
                                    rightlineno=i + 1,
                                    rightline=rightlines[i])
 
     s = difflib.SequenceMatcher(None, leftlines, rightlines)
     if context < 0:
         yield tmpl('comparisonblock', lines=getblock(s.get_opcodes()))
     else:
         for oc in s.get_grouped_opcodes(n=context):
             yield tmpl('comparisonblock', lines=getblock(oc))
 
 def diffstatgen(ctx, basectx):
     '''Generator function that provides the diffstat data.'''
 
     stats = patch.diffstatdata(util.iterlines(ctx.diff(basectx)))
     maxname, maxtotal, addtotal, removetotal, binary = patch.diffstatsum(stats)
     while True:
         yield stats, maxname, maxtotal, addtotal, removetotal, binary
 
 def diffsummary(statgen):
     '''Return a short summary of the diff.'''
 
     stats, maxname, maxtotal, addtotal, removetotal, binary = next(statgen)
     return _(' %d files changed, %d insertions(+), %d deletions(-)\n') % (
              len(stats), addtotal, removetotal)
 
 def diffstat(tmpl, ctx, statgen, parity):
     '''Return a diffstat template for each file in the diff.'''
 
     stats, maxname, maxtotal, addtotal, removetotal, binary = next(statgen)
     files = ctx.files()
 
     def pct(i):
         if maxtotal == 0:
             return 0
         return (float(i) / maxtotal) * 100
 
     fileno = 0
     for filename, adds, removes, isbinary in stats:
         template = filename in files and 'diffstatlink' or 'diffstatnolink'
         total = adds + removes
         fileno += 1
         yield tmpl(template, node=ctx.hex(), file=filename, fileno=fileno,
                    total=total, addpct=pct(adds), removepct=pct(removes),
                    parity=next(parity))
 
 class sessionvars(object):
     def __init__(self, vars, start='?'):
         self.start = start
         self.vars = vars
     def __getitem__(self, key):
         return self.vars[key]
     def __setitem__(self, key, value):
         self.vars[key] = value
     def __copy__(self):
         return sessionvars(copy.copy(self.vars), self.start)
     def __iter__(self):
         separator = self.start
         for key, value in sorted(self.vars.iteritems()):
             yield {'name': key, 'value': str(value), 'separator': separator}
             separator = '&'
 
 class wsgiui(uimod.ui):
     # default termwidth breaks under mod_wsgi
     def termwidth(self):
         return 80
 
 def getwebsubs(repo):
     websubtable = []
     websubdefs = repo.ui.configitems('websub')
     # we must maintain interhg backwards compatibility
     websubdefs += repo.ui.configitems('interhg')
     for key, pattern in websubdefs:
         # grab the delimiter from the character after the "s"
         unesc = pattern[1]
         delim = re.escape(unesc)
 
         # identify portions of the pattern, taking care to avoid escaped
         # delimiters. the replace format and flags are optional, but
         # delimiters are required.
         match = re.match(
             r'^s%s(.+)(?:(?<=\\\\)|(?<!\\))%s(.*)%s([ilmsux])*$'
             % (delim, delim, delim), pattern)
         if not match:
             repo.ui.warn(_("websub: invalid pattern for %s: %s\n")
                               % (key, pattern))
             continue
 
         # we need to unescape the delimiter for regexp and format
         delim_re = re.compile(r'(?<!\\)\\%s' % delim)
         regexp = delim_re.sub(unesc, match.group(1))
         format = delim_re.sub(unesc, match.group(2))
 
         # the pattern allows for 6 regexp flags, so set them if necessary
         flagin = match.group(3)
         flags = 0
         if flagin:
             for flag in flagin.upper():
                 flags |= re.__dict__[flag]
 
         try:
             regexp = re.compile(regexp, flags)
             websubtable.append((regexp, format))
         except re.error:
             repo.ui.warn(_("websub: invalid regexp for %s: %s\n")
                          % (key, regexp))
     return websubtable

mercurial/mdiff.py

@@ -1,459 +1,484 b''
 # mdiff.py - diff and patch routines for mercurial
 #
 # Copyright 2005, 2006 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import re
 import struct
 import zlib
 
 from .i18n import _
 from . import (
     base85,
     bdiff,
     error,
     mpatch,
     pycompat,
     util,
 )
 
 def splitnewlines(text):
     '''like str.splitlines, but only split on newlines.'''
     lines = [l + '\n' for l in text.split('\n')]
     if lines:
         if lines[-1] == '\n':
             lines.pop()
         else:
             lines[-1] = lines[-1][:-1]
     return lines
 
 class diffopts(object):
     '''context is the number of context lines
     text treats all files as text
     showfunc enables diff -p output
     git enables the git extended patch format
     nodates removes dates from diff headers
     nobinary ignores binary files
     noprefix disables the 'a/' and 'b/' prefixes (ignored in plain mode)
     ignorews ignores all whitespace changes in the diff
     ignorewsamount ignores changes in the amount of whitespace
     ignoreblanklines ignores changes whose lines are all blank
     upgrade generates git diffs to avoid data loss
     '''
 
     defaults = {
         'context': 3,
         'text': False,
         'showfunc': False,
         'git': False,
         'nodates': False,
         'nobinary': False,
         'noprefix': False,
         'index': 0,
         'ignorews': False,
         'ignorewsamount': False,
         'ignoreblanklines': False,
         'upgrade': False,
         'showsimilarity': False,
         }
 
     def __init__(self, **opts):
         opts = pycompat.byteskwargs(opts)
         for k in self.defaults.keys():
             v = opts.get(k)
             if v is None:
                 v = self.defaults[k]
             setattr(self, k, v)
 
         try:
             self.context = int(self.context)
         except ValueError:
             raise error.Abort(_('diff context lines count must be '
                                'an integer, not %r') % self.context)
 
     def copy(self, **kwargs):
         opts = dict((k, getattr(self, k)) for k in self.defaults)
         opts.update(kwargs)
         return diffopts(**opts)
 
 defaultopts = diffopts()
 
 def wsclean(opts, text, blank=True):
     if opts.ignorews:
         text = bdiff.fixws(text, 1)
     elif opts.ignorewsamount:
         text = bdiff.fixws(text, 0)
     if blank and opts.ignoreblanklines:
         text = re.sub('\n+', '\n', text).strip('\n')
     return text
 
 def splitblock(base1, lines1, base2, lines2, opts):
     # The input lines matches except for interwoven blank lines. We
     # transform it into a sequence of matching blocks and blank blocks.
     lines1 = [(wsclean(opts, l) and 1 or 0) for l in lines1]
     lines2 = [(wsclean(opts, l) and 1 or 0) for l in lines2]
     s1, e1 = 0, len(lines1)
     s2, e2 = 0, len(lines2)
     while s1 < e1 or s2 < e2:
         i1, i2, btype = s1, s2, '='
         if (i1 >= e1 or lines1[i1] == 0
             or i2 >= e2 or lines2[i2] == 0):
             # Consume the block of blank lines
             btype = '~'
             while i1 < e1 and lines1[i1] == 0:
                 i1 += 1
             while i2 < e2 and lines2[i2] == 0:
                 i2 += 1
         else:
             # Consume the matching lines
             while i1 < e1 and lines1[i1] == 1 and lines2[i2] == 1:
                 i1 += 1
                 i2 += 1
         yield [base1 + s1, base1 + i1, base2 + s2, base2 + i2], btype
         s1 = i1
         s2 = i2
 
+def hunkinrange(hunk, linerange):
+    """Return True if `hunk` defined as (start, length) is in `linerange`
+    defined as (lowerbound, upperbound).
+
+    >>> hunkinrange((5, 10), (2, 7))
+    True
+    >>> hunkinrange((5, 10), (6, 12))
+    True
+    >>> hunkinrange((5, 10), (13, 17))
+    True
+    >>> hunkinrange((5, 10), (3, 17))
+    True
+    >>> hunkinrange((5, 10), (1, 3))
+    False
+    >>> hunkinrange((5, 10), (18, 20))
+    False
+    >>> hunkinrange((5, 10), (1, 5))
+    False
+    >>> hunkinrange((5, 10), (15, 27))
+    False
+    """
+    start, length = hunk
+    lowerbound, upperbound = linerange
+    return lowerbound < start + length and start < upperbound
+
 def blocksinrange(blocks, rangeb):
     """filter `blocks` like (a1, a2, b1, b2) from items outside line range
     `rangeb` from ``(b1, b2)`` point of view.
 
     Return `filteredblocks, rangea` where:
 
     * `filteredblocks` is list of ``block = (a1, a2, b1, b2), stype`` items of
       `blocks` that are inside `rangeb` from ``(b1, b2)`` point of view; a
       block ``(b1, b2)`` being inside `rangeb` if
       ``rangeb[0] < b2 and b1 < rangeb[1]``;
     * `rangea` is the line range w.r.t. to ``(a1, a2)`` parts of `blocks`.
     """
     lbb, ubb = rangeb
     lba, uba = None, None
     filteredblocks = []
     for block in blocks:
         (a1, a2, b1, b2), stype = block
         if lbb >= b1 and ubb <= b2 and stype == '=':
             # rangeb is within a single "=" hunk, restrict back linerange1
             # by offsetting rangeb
             lba = lbb - b1 + a1
             uba = ubb - b1 + a1
         else:
             if b1 <= lbb < b2:
                 if stype == '=':
                     lba = a2 - (b2 - lbb)
                 else:
                     lba = a1
             if b1 < ubb <= b2:
                 if stype == '=':
                     uba = a1 + (ubb - b1)
                 else:
                     uba = a2
-        if lbb < b2 and b1 < ubb:
+        if hunkinrange((b1, (b2 - b1)), rangeb):
             filteredblocks.append(block)
     if lba is None or uba is None or uba < lba:
         raise error.Abort(_('line range exceeds file size'))
     return filteredblocks, (lba, uba)
 
 def allblocks(text1, text2, opts=None, lines1=None, lines2=None):
     """Return (block, type) tuples, where block is an mdiff.blocks
     line entry. type is '=' for blocks matching exactly one another
     (bdiff blocks), '!' for non-matching blocks and '~' for blocks
     matching only after having filtered blank lines.
     line1 and line2 are text1 and text2 split with splitnewlines() if
     they are already available.
     """
     if opts is None:
         opts = defaultopts
     if opts.ignorews or opts.ignorewsamount:
         text1 = wsclean(opts, text1, False)
         text2 = wsclean(opts, text2, False)
     diff = bdiff.blocks(text1, text2)
     for i, s1 in enumerate(diff):
         # The first match is special.
         # we've either found a match starting at line 0 or a match later
         # in the file.  If it starts later, old and new below will both be
         # empty and we'll continue to the next match.
         if i > 0:
             s = diff[i - 1]
         else:
             s = [0, 0, 0, 0]
         s = [s[1], s1[0], s[3], s1[2]]
 
         # bdiff sometimes gives huge matches past eof, this check eats them,
         # and deals with the special first match case described above
         if s[0] != s[1] or s[2] != s[3]:
             type = '!'
             if opts.ignoreblanklines:
                 if lines1 is None:
                     lines1 = splitnewlines(text1)
                 if lines2 is None:
                     lines2 = splitnewlines(text2)
                 old = wsclean(opts, "".join(lines1[s[0]:s[1]]))
                 new = wsclean(opts, "".join(lines2[s[2]:s[3]]))
                 if old == new:
                     type = '~'
             yield s, type
         yield s1, '='
 
 def unidiff(a, ad, b, bd, fn1, fn2, opts=defaultopts):
     """Return a unified diff as a (headers, hunks) tuple.
 
     If the diff is not null, `headers` is a list with unified diff header
     lines "--- <original>" and "+++ <new>" and `hunks` is a generator yielding
     (hunkrange, hunklines) coming from _unidiff().
     Otherwise, `headers` and `hunks` are empty.
     """
     def datetag(date, fn=None):
         if not opts.git and not opts.nodates:
             return '\t%s' % date
         if fn and ' ' in fn:
             return '\t'
         return ''
 
     sentinel = [], ()
     if not a and not b:
         return sentinel
 
     if opts.noprefix:
         aprefix = bprefix = ''
     else:
         aprefix = 'a/'
         bprefix = 'b/'
 
     epoch = util.datestr((0, 0))
 
     fn1 = util.pconvert(fn1)
     fn2 = util.pconvert(fn2)
 
     def checknonewline(lines):
         for text in lines:
             if text[-1:] != '\n':
                 text += "\n\ No newline at end of file\n"
             yield text
 
     if not opts.text and (util.binary(a) or util.binary(b)):
         if a and b and len(a) == len(b) and a == b:
             return sentinel
         headerlines = []
         hunks = (None, ['Binary file %s has changed\n' % fn1]),
     elif not a:
         b = splitnewlines(b)
         if a is None:
             l1 = '--- /dev/null%s' % datetag(epoch)
         else:
             l1 = "--- %s%s%s" % (aprefix, fn1, datetag(ad, fn1))
         l2 = "+++ %s%s" % (bprefix + fn2, datetag(bd, fn2))
         headerlines = [l1, l2]
         size = len(b)
         hunkrange = (0, 0, 1, size)
         hunklines = ["@@ -0,0 +1,%d @@\n" % size] + ["+" + e for e in b]
         hunks = (hunkrange, checknonewline(hunklines)),
     elif not b:
         a = splitnewlines(a)
         l1 = "--- %s%s%s" % (aprefix, fn1, datetag(ad, fn1))
         if b is None:
             l2 = '+++ /dev/null%s' % datetag(epoch)
         else:
             l2 = "+++ %s%s%s" % (bprefix, fn2, datetag(bd, fn2))
         headerlines = [l1, l2]
         size = len(a)
         hunkrange = (1, size, 0, 0)
         hunklines = ["@@ -1,%d +0,0 @@\n" % size] + ["-" + e for e in a]
         hunks = (hunkrange, checknonewline(hunklines)),
     else:
         diffhunks = _unidiff(a, b, opts=opts)
         try:
             hunkrange, hunklines = next(diffhunks)
         except StopIteration:
             return sentinel
 
         headerlines = [
             "--- %s%s%s" % (aprefix, fn1, datetag(ad, fn1)),
             "+++ %s%s%s" % (bprefix, fn2, datetag(bd, fn2)),
         ]
         def rewindhunks():
             yield hunkrange, checknonewline(hunklines)
             for hr, hl in diffhunks:
                 yield hr, checknonewline(hl)
 
         hunks = rewindhunks()
 
     return headerlines, hunks
 
 def _unidiff(t1, t2, opts=defaultopts):
     """Yield hunks of a headerless unified diff from t1 and t2 texts.
 
     Each hunk consists of a (hunkrange, hunklines) tuple where `hunkrange` is a
     tuple (s1, l1, s2, l2) representing the range information of the hunk to
     form the '@@ -s1,l1 +s2,l2 @@' header and `hunklines` is a list of lines
     of the hunk combining said header followed by line additions and
     deletions.
     """
     l1 = splitnewlines(t1)
     l2 = splitnewlines(t2)
     def contextend(l, len):
         ret = l + opts.context
         if ret > len:
             ret = len
         return ret
 
     def contextstart(l):
         ret = l - opts.context
         if ret < 0:
             return 0
         return ret
 
     lastfunc = [0, '']
     def yieldhunk(hunk):
         (astart, a2, bstart, b2, delta) = hunk
         aend = contextend(a2, len(l1))
         alen = aend - astart
         blen = b2 - bstart + aend - a2
 
         func = ""
         if opts.showfunc:
             lastpos, func = lastfunc
             # walk backwards from the start of the context up to the start of
             # the previous hunk context until we find a line starting with an
             # alphanumeric char.
             for i in xrange(astart - 1, lastpos - 1, -1):
                 if l1[i][0].isalnum():
                     func = ' ' + l1[i].rstrip()[:40]
                     lastfunc[1] = func
                     break
             # by recording this hunk's starting point as the next place to
             # start looking for function lines, we avoid reading any line in
             # the file more than once.
             lastfunc[0] = astart
 
         # zero-length hunk ranges report their start line as one less
         if alen:
             astart += 1
         if blen:
             bstart += 1
 
         hunkrange = astart, alen, bstart, blen
         hunklines = (
             ["@@ -%d,%d +%d,%d @@%s\n" % (hunkrange + (func,))]
             + delta
             + [' ' + l1[x] for x in xrange(a2, aend)]
         )
         yield hunkrange, hunklines
 
     # bdiff.blocks gives us the matching sequences in the files.  The loop
     # below finds the spaces between those matching sequences and translates
     # them into diff output.
     #
     hunk = None
     ignoredlines = 0
     for s, stype in allblocks(t1, t2, opts, l1, l2):
         a1, a2, b1, b2 = s
         if stype != '!':
             if stype == '~':
                 # The diff context lines are based on t1 content. When
                 # blank lines are ignored, the new lines offsets must
                 # be adjusted as if equivalent blocks ('~') had the
                 # same sizes on both sides.
                 ignoredlines += (b2 - b1) - (a2 - a1)
             continue
         delta = []
         old = l1[a1:a2]
         new = l2[b1:b2]
 
         b1 -= ignoredlines
         b2 -= ignoredlines
         astart = contextstart(a1)
         bstart = contextstart(b1)
         prev = None
         if hunk:
             # join with the previous hunk if it falls inside the context
             if astart < hunk[1] + opts.context + 1:
                 prev = hunk
                 astart = hunk[1]
                 bstart = hunk[3]
             else:
                 for x in yieldhunk(hunk):
                     yield x
         if prev:
             # we've joined the previous hunk, record the new ending points.
             hunk[1] = a2
             hunk[3] = b2
             delta = hunk[4]
         else:
             # create a new hunk
             hunk = [astart, a2, bstart, b2, delta]
 
         delta[len(delta):] = [' ' + x for x in l1[astart:a1]]
         delta[len(delta):] = ['-' + x for x in old]
         delta[len(delta):] = ['+' + x for x in new]
 
     if hunk:
         for x in yieldhunk(hunk):
             yield x
 
 def b85diff(to, tn):
     '''print base85-encoded binary diff'''
     def fmtline(line):
         l = len(line)
         if l <= 26:
             l = chr(ord('A') + l - 1)
         else:
             l = chr(l - 26 + ord('a') - 1)
         return '%c%s\n' % (l, base85.b85encode(line, True))
 
     def chunk(text, csize=52):
         l = len(text)
         i = 0
         while i < l:
             yield text[i:i + csize]
             i += csize
 
     if to is None:
         to = ''
     if tn is None:
         tn = ''
 
     if to == tn:
         return ''
 
     # TODO: deltas
     ret = []
     ret.append('GIT binary patch\n')
     ret.append('literal %s\n' % len(tn))
     for l in chunk(zlib.compress(tn)):
         ret.append(fmtline(l))
     ret.append('\n')
 
     return ''.join(ret)
 
 def patchtext(bin):
     pos = 0
     t = []
     while pos < len(bin):
         p1, p2, l = struct.unpack(">lll", bin[pos:pos + 12])
         pos += 12
         t.append(bin[pos:pos + l])
         pos += l
     return "".join(t)
 
 def patch(a, bin):
     if len(a) == 0:
         # skip over trivial delta header
         return util.buffer(bin, 12)
     return mpatch.patches(a, [bin])
 
 # similar to difflib.SequenceMatcher.get_matching_blocks
 def get_matching_blocks(a, b):
     return [(d[0], d[2], d[1] - d[0]) for d in bdiff.blocks(a, b)]
 
 def trivialdiffheader(length):
     return struct.pack(">lll", 0, 0, length) if length else ''
 
 def replacediffheader(oldlen, newlen):
     return struct.pack(">lll", 0, oldlen, newlen)
 
 patches = mpatch.patches
 patchedsize = mpatch.patchedsize
 textdiff = bdiff.bdiff

tests/test-doctest.py

@@ -1,55 +1,56 b''
 # this is hack to make sure no escape characters are inserted into the output
 
 from __future__ import absolute_import
 
 import doctest
 import os
 import sys
 
 ispy3 = (sys.version_info[0] >= 3)
 
 if 'TERM' in os.environ:
     del os.environ['TERM']
 
 # TODO: migrate doctests to py3 and enable them on both versions
 def testmod(name, optionflags=0, testtarget=None, py2=True, py3=False):
     if not (not ispy3 and py2 or ispy3 and py3):
         return
     __import__(name)
     mod = sys.modules[name]
     if testtarget is not None:
         mod = getattr(mod, testtarget)
     doctest.testmod(mod, optionflags=optionflags)
 
 testmod('mercurial.changegroup')
 testmod('mercurial.changelog')
 testmod('mercurial.color')
 testmod('mercurial.config')
 testmod('mercurial.dagparser', optionflags=doctest.NORMALIZE_WHITESPACE)
 testmod('mercurial.dispatch')
 testmod('mercurial.encoding')
 testmod('mercurial.formatter')
 testmod('mercurial.hg')
 testmod('mercurial.hgweb.hgwebdir_mod')
 testmod('mercurial.match')
+testmod('mercurial.mdiff')
 testmod('mercurial.minirst')
 testmod('mercurial.patch')
 testmod('mercurial.pathutil')
 testmod('mercurial.parser')
 testmod('mercurial.pycompat', py3=True)
 testmod('mercurial.revsetlang')
 testmod('mercurial.smartset')
 testmod('mercurial.store')
 testmod('mercurial.subrepo')
 testmod('mercurial.templatefilters')
 testmod('mercurial.templater')
 testmod('mercurial.ui')
 testmod('mercurial.url')
 testmod('mercurial.util')
 testmod('mercurial.util', testtarget='platform')
 testmod('hgext.convert.convcmd')
 testmod('hgext.convert.cvsps')
 testmod('hgext.convert.filemap')
 testmod('hgext.convert.p4')
 testmod('hgext.convert.subversion')
 testmod('hgext.mq')