X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Fparser_tools.py;h=6d64a793ce2e3861b5a922e3930f3bd0351cc005;hb=a84a98b54561044b5902dfce39bc31082877fec1;hp=d0466df703ded682baecbe1b2bfd5c7575404021;hpb=b1c11faf7e29bf24d2315fd04a2515a2fd3c126a;p=lyx.git

diff --git a/lib/lyx2lyx/parser_tools.py b/lib/lyx2lyx/parser_tools.py
index d0466df703..6d64a793ce 100644
--- a/lib/lyx2lyx/parser_tools.py
+++ b/lib/lyx2lyx/parser_tools.py
@@ -1,6 +1,6 @@
 # This file is part of lyx2lyx
 # -*- coding: utf-8 -*-
-# Copyright (C) 2002-2010 Dekel Tsur <dekel@lyx.org>, 
+# Copyright (C) 2002-2011 Dekel Tsur <dekel@lyx.org>, 
 # José Matos <jamatos@lyx.org>, Richard Heck <rgheck@comcast.net>
 #
 # This program is free software; you can redistribute it and/or
@@ -15,7 +15,7 @@
 #
 # You should have received a copy of the GNU General Public License
 # along with this program; if not, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
 
 
 ''' 
@@ -23,23 +23,27 @@ This modules offer several free functions to help parse lines.
 More documentaton is below, but here is a quick guide to what 
 they do. Optional arguments are marked by brackets.
 
-find_token(lines, token, start[, end[, exact]]):
+find_token(lines, token, start[, end[, ignorews]]):
   Returns the first line i, start <= i < end, on which
   token is found at the beginning. Returns -1 if not 
-  found. If exact is (given and) True, then differences
-  in whitespace do not count.
+  found. 
+  If ignorews is (given and) True, then differences
+  in whitespace do not count, except that there must be no 
+  extra whitespace following token itself.
 
 find_token_exact(lines, token, start[, end]):
-  Badly named. As find_token, but with ignorews True.
+  As find_token, but with ignorews True.
 
 find_tokens(lines, tokens, start[, end[, ignorews]]):
   Returns the first line i, start <= i < end, on which
   oen of the tokens in tokens is found at the beginning. 
-  Returns -1 if not found. If ignorews is (given and) True, 
-  then differences in whitespace do not count.
+  Returns -1 if not found. 
+  If ignorews is (given and) True, then differences
+  in whitespace do not count, except that there must be no 
+  extra whitespace following token itself.
 
 find_tokens_exact(lines, token, start[, end]):
-  Badly named. As find_tokens, but with ignorews True.
+  As find_tokens, but with ignorews True.
   
 find_token_backwards(lines, token, start):
 find_tokens_backwards(lines, tokens, start):
@@ -65,11 +69,81 @@ get_quoted_value(lines, token, start[, end[, default]):
   value, if they are present. So use this one for cases
   where the value is normally quoted.
 
-del_token(lines, token, start, end):
+get_option_value(line, option):
+  This assumes we have a line with something like:
+      option="value"
+  and returns value. Returns "" if not found.
+
+del_token(lines, token, start[, end]):
   Like find_token, but deletes the line if it finds one.
   Returns True if a line got deleted, otherwise False.
+
+find_beginning_of(lines, i, start_token, end_token):
+  Here, start_token and end_token are meant to be a matching 
+  pair, like "\begin_layout" and "\end_layout". We look for 
+  the start_token that pairs with the end_token that occurs
+  on or after line i. Returns -1 if not found.
+  So, in the layout case, this would find the \begin_layout 
+  for the layout line i is in. 
+  Example:
+    ec = find_token(document.body, "</cell", i)
+    bc = find_beginning_of(document.body, ec, \
+        "<cell", "</cell")
+  Now, assuming no -1s, bc-ec wraps the cell for line i.
+
+find_end_of(lines, i, start_token, end_token):
+  Like find_beginning_of, but looking for the matching 
+  end_token. This might look like:
+    bc = find_token_(document.body, "<cell", i)
+    ec = find_end_of(document.body, bc,  "<cell", "</cell")
+  Now, assuming no -1s, bc-ec wrap the next cell.
+
+find_end_of_inset(lines, i):
+  Specialization of find_end_of for insets.
+
+find_end_of_layout(lines, i):
+  Specialization of find_end_of for layouts.
+
+is_in_inset(lines, i, inset):
+  Checks if line i is in an inset of the given type.
+  If so, returns starting and ending lines. Otherwise, 
+  returns False.
+  Example:
+    is_in_inset(document.body, i, "\\begin_inset Tabular")
+  returns False unless i is within a table. If it is, then
+  it returns the line on which the table begins and the one
+  on which it ends. Note that this pair will evaulate to
+  boolean True, so
+    if is_in_inset(...):
+  will do what you expect.
+
+get_containing_inset(lines, i):
+  Finds out what kind of inset line i is within. Returns a 
+  list containing what follows \begin_inset on the the line 
+  on which the inset begins, plus the starting and ending line.
+  Returns False on any kind of error or if it isn't in an inset.
+  So get_containing_inset(document.body, i) might return:
+    ("CommandInset ref", 300, 306)
+  if i is within an InsetRef beginning on line 300 and ending
+  on line 306.
+
+get_containing_layout(lines, i):
+  As get_containing_inset, but for layout.
+
+
+find_nonempty_line(lines, start[, end):
+  Finds the next non-empty line.
+
+check_token(line, token):
+  Does line begin with token?
+
+is_nonempty_line(line):
+  Does line contain something besides whitespace?
+
 '''
 
+import re
+
 # Utilities for one line
 def check_token(line, token):
     """ check_token(line, token) -> bool
@@ -89,14 +163,15 @@ def is_nonempty_line(line):
 
 
 # Utilities for a list of lines
-def find_token(lines, token, start, end = 0, exact = False):
-    """ find_token(lines, token, start[[, end], exact]) -> int
+def find_token(lines, token, start, end = 0, ignorews = False):
+    """ find_token(lines, token, start[[, end], ignorews]) -> int
 
     Return the lowest line where token is found, and is the first
     element, in lines[start, end].
     
-    If exact is True (default is False), then differences in
-    whitespace are ignored.
+    If ignorews is True (default is False), then differences in
+    whitespace are ignored, except that there must be no extra
+    whitespace following token itself.
 
     Return -1 on failure."""
 
@@ -104,7 +179,7 @@ def find_token(lines, token, start, end = 0, exact = False):
         end = len(lines)
     m = len(token)
     for i in xrange(start, end):
-        if exact:
+        if ignorews:
             x = lines[i].split()
             y = token.split()
             if len(x) < len(y):
@@ -121,19 +196,19 @@ def find_token_exact(lines, token, start, end = 0):
     return find_token(lines, token, start, end, True)
 
 
-def find_tokens(lines, tokens, start, end = 0, exact = False):
-    """ find_tokens(lines, tokens, start[[, end], exact]) -> int
+def find_tokens(lines, tokens, start, end = 0, ignorews = False):
+    """ find_tokens(lines, tokens, start[[, end], ignorews]) -> int
 
     Return the lowest line where one token in tokens is found, and is
     the first element, in lines[start, end].
 
     Return -1 on failure."""
-    if end == 0:
+    if end == 0 or end > len(lines):
         end = len(lines)
 
     for i in xrange(start, end):
         for token in tokens:
-            if exact:
+            if ignorews:
                 x = lines[i].split()
                 y = token.split()
                 if len(x) < len(y):
@@ -158,7 +233,7 @@ def find_re(lines, rexp, start, end = 0):
 
     Return -1 on failure."""
 
-    if end == 0:
+    if end == 0 or end > len(lines):
         end = len(lines)
     for i in xrange(start, end):
         if rexp.match(lines[i]):
@@ -231,7 +306,25 @@ def get_quoted_value(lines, token, start, end = 0, default = ""):
     return val.strip('"')
 
 
-def del_token(lines, token, start, end):
+def get_option_value(line, option):
+    rx = option + '\s*=\s*"([^"]+)"'
+    rx = re.compile(rx)
+    m = rx.search(line)
+    if not m:
+      return ""
+    return m.group(1)
+
+
+def set_option_value(line, option, value):
+    rx = '(' + option + '\s*=\s*")[^"]+"'
+    rx = re.compile(rx)
+    m = rx.search(line)
+    if not m:
+        return line
+    return re.sub(rx, '\g<1>' + value + '"', line)
+
+
+def del_token(lines, token, start, end = 0):
     """ del_token(lines, token, start, end) -> int
 
     Find the first line in lines where token is the first element 
@@ -327,12 +420,16 @@ def get_containing_inset(lines, i):
   on which the inset begins, plus the starting and ending line.
   Returns False on any kind of error or if it isn't in an inset.
   '''
-  stins = find_token_backwards(lines, i, "\\begin_inset")
-  if stins == -1:
-      return False
-  endins = find_end_of_inset(lines, stins)
-  if endins < i:
-      return False
+  j = i
+  while True:
+      stins = find_token_backwards(lines, "\\begin_inset", j)
+      if stins == -1:
+          return False
+      endins = find_end_of_inset(lines, stins)
+      if endins > j:
+          break
+      j = stins - 1
+  
   inset = get_value(lines, "\\begin_inset", stins)
   if inset == "":
       # shouldn't happen
@@ -347,14 +444,18 @@ def get_containing_layout(lines, i):
   on which the layout begins, plus the starting and ending line.
   Returns False on any kind of error.
   '''
-  stins = find_token_backwards(lines, i, "\\begin_layout")
-  if stins == -1:
-      return False
-  endins = find_end_of_layout(lines, stins)
-  if endins < i:
-      return False
-  lay = get_value(lines, "\\begin_layout", stins)
+  j = i
+  while True:
+      stlay = find_token_backwards(lines, "\\begin_layout", j)
+      if stlay == -1:
+          return False
+      endlay = find_end_of_layout(lines, stlay)
+      if endlay > i:
+          break
+      j = stlay - 1
+  
+  lay = get_value(lines, "\\begin_layout", stlay)
   if lay == "":
       # shouldn't happen
       return False
-  return (lay, stins, endins)
+  return (lay, stlay, endlay)