X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Fparser_tools.py;h=3b1322974e6878e058be8533b7ca8c73c60cca63;hb=9da74fe2078e24e1e7891784ecbfe33ff77e7f85;hp=8db73c615a62af449561467fb127b841454e15b4;hpb=e0227fa3968d0e55f145fbb6937476c2f9f66bd4;p=lyx.git

diff --git a/lib/lyx2lyx/parser_tools.py b/lib/lyx2lyx/parser_tools.py
index 8db73c615a..3b1322974e 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
@@ -15,9 +15,144 @@
 #
 # 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.
-
-" This modules offer several free functions to help parse lines. "
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
+
+
+''' 
+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[, ignorews]]):
+  Returns the first line i, start <= i < end, on which
+  token is found at the beginning. 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_token_exact(lines, token, start[, end]):
+  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, except that there must be no 
+  extra whitespace following token itself.
+
+find_tokens_exact(lines, token, start[, end]):
+  As find_tokens, but with ignorews True.
+  
+find_token_backwards(lines, token, start):
+find_tokens_backwards(lines, tokens, start):
+  As before, but look backwards.
+
+find_re(lines, rexp, start[, end]):
+  As find_token, but rexp is a regular expression object,
+  so it has to be passed as e.g.: re.compile(r'...').
+
+get_value(lines, token, start[, end[, default]):
+  Similar to find_token, but it returns what follows the 
+  token on the found line. Example:
+    get_value(document.header, "\use_xetex", 0)
+  will find a line like:
+    \use_xetex true
+  and, in that case, return "true". (Note that whitespace
+  is stripped.) The final argument, default, defaults to "", 
+  and is what is returned if we do not find anything. So you
+  can use that to set a default.
+  
+get_quoted_value(lines, token, start[, end[, default]):
+  Similar to get_value, but it will strip quotes off the
+  value, if they are present. So use this one for cases
+  where the value is normally quoted.
+
+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.
+
+find_end_of_sequence(lines, i):
+  Find the end of the sequence of layouts of the same kind.
+  Considers nesting. If the last paragraph in sequence is nested,
+  the position of the last \end_deeper is returned, else
+  the position of the last \end_layout.
+
+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 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. Additionally returns the
+  position of real paragraph start (after par params) as 4th value.
+
+
+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?
+
+count_pars_in_inset(lines, i):
+  Counts the paragraphs inside an inset.
+
+'''
+
+import re
 
 # Utilities for one line
 def check_token(line, token):
@@ -38,11 +173,15 @@ 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 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."""
 
@@ -50,7 +189,7 @@ 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):
@@ -67,19 +206,19 @@ 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):
@@ -104,7 +243,7 @@ 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]):
@@ -143,55 +282,70 @@ def find_tokens_backwards(lines, tokens, start):
 
 
 def get_value(lines, token, start, end = 0, default = ""):
-    """ get_value(lines, token, start[[, end], default]) -> list of strings
-
-    Return tokens after token for the first line, in lines, where
-    token is the first element."""
-
-    i = find_token_exact(lines, token, start, end)
-    if i == -1:
-        return default
-    if len(lines[i].split()) > 1:
-        return lines[i].split()[1]
-    else:
-        return default
-
-
-def get_value_string(lines, token, start, end = 0, trim = False, default = ""):
-    """ get_value_string(lines, token, start[[, end], trim, default]) -> string
+    """ get_value(lines, token, start[[, end], default]) -> string
 
-    Return tokens after token as string, in lines, where
-    token is the first element. When trim is used, the first and last character
-    of the string is trimmed."""
+    Find the next line that looks like:
+      token followed by other stuff
+    Returns "followed by other stuff" with leading and trailing
+    whitespace removed.
+    """
 
     i = find_token_exact(lines, token, start, end)
     if i == -1:
         return default
-    if len(lines[i].split()) > 1:
-        for k in range (0, len(lines[i])):
-            if lines[i][k] == ' ':
-                if trim ==False:
-                    return lines[i][k+1:len(lines[i])]
-                else:
-                    return lines[i][k+2:len(lines[i])-1]
-    else:
-        return default
-
-
-def del_token(lines, token, start, end):
+    l = lines[i].split(None, 1)
+    if len(l) > 1:
+        return l[1].strip()
+    return default
+
+
+def get_quoted_value(lines, token, start, end = 0, default = ""):
+    """ get_quoted_value(lines, token, start[[, end], default]) -> string
+
+    Find the next line that looks like:
+      token "followed by other stuff"
+    Returns "followed by other stuff" with leading and trailing
+    whitespace and quotes removed. If there are no quotes, that is OK too.
+    So use get_value to preserve possible quotes, this one to remove them,
+    if they are there.
+    Note that we will NOT strip quotes from default!
+    """
+    val = get_value(lines, token, start, end, "")
+    if not val:
+      return default
+    return val.strip('"')
+
+
+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 lower line in lines where token is the first element and
-    delete that line.
-
-    Returns the number of lines remaining."""
+    Find the first line in lines where token is the first element 
+    and delete that line. Returns True if we deleted a line, False
+    if we did not."""
 
     k = find_token_exact(lines, token, start, end)
     if k == -1:
-        return end
-    else:
-        del lines[k]
-        return end - 1
+        return False
+    del lines[k]
+    return True
 
 
 def find_beginning_of(lines, i, start_token, end_token):
@@ -244,10 +398,20 @@ def find_end_of_layout(lines, i):
     return find_end_of(lines, i, "\\begin_layout", "\\end_layout")
 
 
-# checks if line i is in the inset e.g., "\\begin_inset CommandInset ref"
-# if so, returns starting and ending lines
-# otherwise, returns (-1, -1)
-def get_containing_inset(lines, i, inset):
+def 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.
+    '''
     defval = (-1, -1)
     stins = find_token_backwards(lines, inset, i)
     if stins == -1:
@@ -257,3 +421,109 @@ def get_containing_inset(lines, i, inset):
     if endins < i:
       return defval
     return (stins, endins)
+
+
+def get_containing_inset(lines, i):
+  ''' 
+  Finds out what kind of inset line i is within. Returns a 
+  list containing (i) what follows \begin_inset on 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.
+  '''
+  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
+      return False
+  return (inset, stins, endins)
+
+
+def get_containing_layout(lines, i):
+  ''' 
+  Finds out what kind of layout line i is within. Returns a 
+  list containing (i) what follows \begin_layout on the line 
+  on which the layout begins, plus the starting and ending line
+  and the start of the apargraph (after all params).
+  Returns False on any kind of error.
+  '''
+  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
+  par_params = ["\\noindent", "\\indent", "\\indent-toggle", "\\leftindent",
+                "\\start_of_appendix", "\\paragraph_spacing single",
+                "\\paragraph_spacing onehalf", "\\paragraph_spacing double",
+                "\\paragraph_spacing other", "\\align", "\\labelwidthstring"]
+  stpar = stlay
+  while True:
+      stpar += 1
+      if lines[stpar] not in par_params:
+          break
+  return (lay, stlay, endlay, stpar)
+
+
+def count_pars_in_inset(lines, i):
+  ''' 
+  Counts the paragraphs within this inset
+  '''
+  ins = get_containing_inset(lines, i)
+  if ins == -1:
+      return -1
+  pars = 0
+  for j in range(ins[1], ins[2]):
+      m = re.match(r'\\begin_layout (.*)', lines[j])
+      if m and get_containing_inset(lines, j)[0] == ins[0]:
+          pars += 1
+  
+  return pars
+
+
+def find_end_of_sequence(lines, i):
+  ''' 
+  Returns the end of a sequence of identical layouts.
+  '''
+  lay = get_containing_layout(lines, i)
+  if lay == False:
+      return -1
+  layout = lay[0]
+  endlay = lay[2]
+  i = endlay
+  while True:
+      m = re.match(r'\\begin_layout (.*)', lines[i])
+      if m and m.group(1) != layout:
+          return endlay
+      elif lines[i] == "\\begin_deeper":
+          j = find_end_of(lines, i, "\\begin_deeper", "\\end_deeper")
+          if j != -1:
+              i = j
+              endlay = j
+              continue
+      if m and m.group(1) == layout:
+          endlay = find_end_of_layout(lines, i)
+          i = endlay
+          continue
+      if i == len(lines) - 1:
+          break
+      i = i + 1
+
+  return endlay
+