X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Flyx_2_0.py;h=2756418f88d09afb0671878977800b6a8f525584;hb=9da74fe2078e24e1e7891784ecbfe33ff77e7f85;hp=4abacaaf2f0f866d1adcf243854ee3058a244798;hpb=a1589fb0910c6895d438c759eaa0d5b5ba6f913c;p=lyx.git

diff --git a/lib/lyx2lyx/lyx_2_0.py b/lib/lyx2lyx/lyx_2_0.py
index 4abacaaf2f..2756418f88 100644
--- a/lib/lyx2lyx/lyx_2_0.py
+++ b/lib/lyx2lyx/lyx_2_0.py
@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 # This file is part of lyx2lyx
 # -*- coding: utf-8 -*-
-# Copyright (C) 2010 The LyX team
+# Copyright (C) 2011 The LyX team
 #
 # This program is free software; you can redistribute it and/or
 # modify it under the terms of the GNU General Public License
@@ -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
 
 """ Convert files to the file format generated by lyx 2.0"""
 
@@ -45,17 +45,6 @@ def remove_option(lines, m, option):
     return True
 
 
-# DO NOT USE THIS ROUTINE ANY MORE. Better yet, replace the uses that
-# have been made of it with uses of put_cmd_in_ert.
-def old_put_cmd_in_ert(string):
-    for rep in unicode_reps:
-        string = string.replace(rep[1], rep[0].replace('\\\\', '\\'))
-    string = string.replace('\\', "\\backslash\n")
-    string = "\\begin_inset ERT\nstatus collapsed\n\\begin_layout Plain Layout\n" \
-      + string + "\n\\end_layout\n\\end_inset"
-    return string
-
-
 ###############################################################################
 ###
 ### Conversion and reversion routines
@@ -214,7 +203,7 @@ def revert_xetex(document):
 
     # 2.) check font settings
     # defaults
-    roman = sans = typew = default
+    roman = sans = typew = "default"
     osf = False
     sf_scale = tt_scale = 100.0
     
@@ -288,11 +277,11 @@ def revert_xetex(document):
             sf += 'Scale=' + str(sf_scale / 100.0) + ','
         sf += 'Mapping=tex-text]{' + sans + '}'
         pretext.append(sf)
-    if typewriter != "default":
+    if typew != "default":
         tw = '\\setmonofont'
         if tt_scale != 100.0:
             tw += '[Scale=' + str(tt_scale / 100.0) + ']'
-        tw += '{' + typewriter + '}'
+        tw += '{' + typew + '}'
         pretext.append(tw)
     if osf:
         pretext.append('\\defaultfontfeatures{Numbers=OldStyle}')
@@ -771,7 +760,7 @@ def revert_author_id(document):
     " Remove the author_id from the \\author definition "
     i = 0
     anum = 0
-    rx = re.compile(r'(\\author)\s+(\d+)\s+(\".*\")\s*(.*)$')
+    rx = re.compile(r'(\\author)\s+(-?\d+)\s+(\".*\")\s*(.*)$')
     idmap = dict()
 
     while True:
@@ -816,6 +805,28 @@ def revert_suppress_date(document):
     del document.header[i]
 
 
+def convert_mhchem(document):
+    "Set mhchem to off for versions older than 1.6.x"
+    if document.start < 277:
+        # LyX 1.5.x and older did never load mhchem.
+        # Therefore we must switch it off: Documents that use mhchem have
+        # a manual \usepackage anyway, and documents not using mhchem but
+        # custom macros with the same names as mhchem commands might get
+        # corrupted if mhchem is automatically loaded.
+        mhchem = 0 # off
+    else:
+        # LyX 1.6.x did always load mhchem automatically.
+        mhchem = 1 # auto
+    i = find_token(document.header, "\\use_esint", 0)
+    if i == -1:
+        # pre-1.5.x document
+        i = find_token(document.header, "\\use_amsmath", 0)
+    if i == -1:
+        document.warning("Malformed LyX document: Could not find amsmath os esint setting.")
+        return
+    document.header.insert(i + 1, "\\use_mhchem %d" % mhchem)
+
+
 def revert_mhchem(document):
     "Revert mhchem loading to preamble code"
 
@@ -972,7 +983,6 @@ def revert_multirow(document):
           numrows = int(numrows)
           numcols = int(numcols)
         except:
-          document.warning(numrows)
           document.warning("Unable to determine rows and columns!")
           begin_table = end_table
           continue
@@ -1549,7 +1559,7 @@ def convert_use_makebox(document):
       i = z
       continue
     document.body.insert(k + 1, "use_makebox 0")
-    i = z + 1
+    i = blay + 1 # not z + 1 (box insets may be nested)
 
 
 def revert_IEEEtran(document):
@@ -1805,9 +1815,18 @@ def revert_flexnames(document):
 
 def convert_mathdots(document):
     " Load mathdots automatically "
-    i = find_token(document.header, "\\use_esint" , 0)
-    if i != -1:
-      document.header.insert(i + 1, "\\use_mathdots 1")
+    i = find_token(document.header, "\\use_mhchem" , 0)
+    if i == -1:
+        i = find_token(document.header, "\\use_esint" , 0)
+    if i == -1:
+        document.warning("Malformed LyX document: Can't find \\use_mhchem.")
+        return;
+    j = find_token(document.preamble, "\\usepackage{mathdots}", 0)
+    if j == -1:
+        document.header.insert(i + 1, "\\use_mathdots 0")
+    else:
+        document.header.insert(i + 1, "\\use_mathdots 2")
+        del document.preamble[j]
 
 
 def revert_mathdots(document):
@@ -1815,7 +1834,7 @@ def revert_mathdots(document):
 
     mathdots = find_token(document.header, "\\use_mathdots" , 0)
     if mathdots == -1:
-      document.warning("No \\usemathdots line. Assuming auto.")
+      document.warning("No \\use_mathdots line. Assuming auto.")
     else:
       val = get_value(document.header, "\\use_mathdots", mathdots)
       del document.header[mathdots]
@@ -1831,7 +1850,7 @@ def revert_mathdots(document):
         return
       if usedots == 2:
         # force load case
-        add_to_preamble(["\\usepackage{mathdots}"])
+        add_to_preamble(document, ["\\usepackage{mathdots}"])
         return
     
     # so we are in the auto case. we want to load mathdots if \iddots is used.
@@ -1954,10 +1973,17 @@ def revert_diagram(document):
     # only need to do it once!
     return
 
+chapters = ("amsbook", "book", "docbook-book", "elsart", "extbook", "extreport", 
+    "jbook", "jreport", "jsbook", "literate-book", "literate-report", "memoir", 
+    "mwbk", "mwrep", "recipebook", "report", "scrbook", "scrreprt", "svmono", 
+    "svmult", "tbook", "treport", "tufte-book")
 
 def convert_bibtex_clearpage(document):
   " insert a clear(double)page bibliographystyle if bibtotoc option is used "
 
+  if document.textclass not in chapters:
+    return
+
   i = find_token(document.header, '\\papersides', 0)
   sides = 0
   if i == -1:
@@ -2047,9 +2073,17 @@ def convert_passthru(document):
           document.warning("Can't find end of layout at line " + str(beg))
           beg += 1
           continue
+
         # we are now going to replace newline insets within this layout
         # by new instances of this layout. so we have repeated layouts
         # instead of newlines.
+
+        # if the paragraph has any customization, however, we do not want to
+        # do the replacement.
+        if document.body[beg + 1].startswith("\\"):
+          beg = end + 1
+          continue
+
         ns = beg
         while True:
           ns = find_token(document.body, "\\begin_inset Newline newline", ns, end)
@@ -2068,7 +2102,8 @@ def convert_passthru(document):
           # do ns properly, too
           newlines = (ne - ns) - len(subst)
           ns += newlines + 2
-          end += newlines + 1
+          end += newlines + 2
+
         # ok, we now want to find out if the next layout is the
         # same as this one. if so, we will insert an extra copy of it
         didit = False
@@ -2159,133 +2194,263 @@ def revert_multirowOffset(document):
     " Revert multirow cells with offset in tables to TeX-code"
     # this routine is the same as the revert_multirow routine except that
     # it checks additionally for the offset
-    i = 0
-    begin_table = 0
-    bottom_index = 0
-    multirowOffset = False
-    # cell type 3 is multirow begin cell
-    i = find_token(document.body, '<cell multirow="3" mroffset=', i)
+
+    # first, let's find out if we need to do anything
+    i = find_token(document.body, '<cell multirow="3" mroffset=', 0)
     if i == -1:
       return
-    # a multirow cell with offset was found
-    multirowOffset = True
+
+    add_to_preamble(document, ["\\usepackage{multirow}"])
+
+    rgx = re.compile(r'mroffset="[^"]+?"')
+    begin_table = 0
+
     while True:
-      # find begin/end of table
-      begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
-      if begin_table == -1:
-          break
-      end_table = find_token(document.body, '</lyxtabular>', begin_table)
-      if end_table == -1:
-          document.warning("Malformed LyX document: Could not find end of table.")
-          break
-      # store the number of rows
-      begin = document.body[begin_table].find('rows="')
-      end = document.body[begin_table].find('" ', begin)
-      rows = document.body[begin_table][begin + 6:end]
-      # store the number of columns
-      begin = document.body[begin_table].find('columns="')
-      end = document.body[begin_table].find('">', begin)
-      columns = document.body[begin_table][begin + 9:end]
-      # find column number of the multirows with offset
-      begin_row = begin_table
-      for row in range(int(rows)):
-        begin_row = find_token(document.body, '<row>', begin_row)
-        begin_cell = begin_row
-        for column in range(int(columns)):            
-          begin_cell = find_token(document.body, '<cell ', begin_cell)
-          i = document.body[begin_cell].find('multirow="3" mroffset=')
-          if i <> -1:
-            # we found a multirow in column number column - 1
-            # store column width
-            begin = document.body[begin_table + 2 + column].find('width="')
-            end = document.body[begin_table + 2 + column].find('">', begin)
-            col_width = document.body[begin_table + 2 + column][begin + 7:end]
+        # find begin/end of table
+        begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
+        if begin_table == -1:
+            break
+        end_table = find_end_of(document.body, begin_table, '<lyxtabular', '</lyxtabular>')
+        if end_table == -1:
+            document.warning("Malformed LyX document: Could not find end of table.")
+            begin_table += 1
+            continue
+        # does this table have multirow?
+        i = find_token(document.body, '<cell multirow="3"', begin_table, end_table)
+        if i == -1:
+            begin_table = end_table
+            continue
+        
+        # store the number of rows and columns
+        numrows = get_option_value(document.body[begin_table], "rows")
+        numcols = get_option_value(document.body[begin_table], "columns")
+        try:
+          numrows = int(numrows)
+          numcols = int(numcols)
+        except:
+          document.warning("Unable to determine rows and columns!")
+          begin_table = end_table
+          continue
+
+        mrstarts = []
+        multirows = []
+        # collect info on rows and columns of this table.
+        begin_row = begin_table
+        for row in range(numrows):
+            begin_row = find_token(document.body, '<row>', begin_row, end_table)
+            if begin_row == -1:
+              document.warning("Can't find row " + str(row + 1))
+              break
+            end_row = find_end_of(document.body, begin_row, '<row>', '</row>')
+            if end_row == -1:
+              document.warning("Can't find end of row " + str(row + 1))
+              break
+            begin_cell = begin_row
+            multirows.append([])
+            for column in range(numcols):            
+                begin_cell = find_token(document.body, '<cell ', begin_cell, end_row)
+                if begin_cell == -1:
+                  document.warning("Can't find column " + str(column + 1) + \
+                    "in row " + str(row + 1))
+                  break
+                # NOTE 
+                # this will fail if someone puts "</cell>" in a cell, but
+                # that seems fairly unlikely.
+                end_cell = find_end_of(document.body, begin_cell, '<cell', '</cell>')
+                if end_cell == -1:
+                  document.warning("Can't find end of column " + str(column + 1) + \
+                    "in row " + str(row + 1))
+                  break
+                multirows[row].append([begin_cell, end_cell, 0])
+                if document.body[begin_cell].find('multirow="3" mroffset=') != -1:
+                  multirows[row][column][2] = 3 # begin multirow
+                  mrstarts.append([row, column])
+                elif document.body[begin_cell].find('multirow="4"') != -1:
+                  multirows[row][column][2] = 4 # in multirow
+                begin_cell = end_cell
+            begin_row = end_row
+        # end of table info collection
+
+        # work from the back to avoid messing up numbering
+        mrstarts.reverse()
+        for m in mrstarts:
+            row = m[0]
+            col = m[1]
+            # get column width
+            col_width = get_option_value(document.body[begin_table + 2 + col], "width")
             # "0pt" means that no width is specified
-            if col_width == "0pt":
+            if not col_width or col_width == "0pt":
               col_width = "*"
             # determine the number of cells that are part of the multirow
-            begin_row2 = begin_table
-            # we have the multirow begin cell as minimal span
-            span = 1
-            # go one row behind the row where we found the multirow begin
-            row2 = -1
-            while row2 < row:
-              begin_row2 = find_token(document.body, '<row>', begin_row2)
-              row2 += 1
-              begin_row2 += 1
-            # step now through all rows until we reach the last table row
-            rows2 = row
-            while row2 < rows:
-              begin_row2 = find_token(document.body, '<row>', begin_row2)
-              begin_cell2 = begin_row2
-              # look now for the column-th cell in the row
-              column2 = 0
-              while column2 < column:
-                begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
-                column2 += 1
-                begin_cell2 += 1
-              # cell type 4 is multirow-part cell
-              begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
-              j = document.body[begin_cell2].find('multirow="4"')
-              if j <> -1:
-                span += 1
+            nummrs = 1
+            for r in range(row + 1, numrows):
+                if multirows[r][col][2] != 4:
+                  break
+                nummrs += 1
+                # take the opportunity to revert this line
+                lineno = multirows[r][col][0]
+                document.body[lineno] = document.body[lineno].\
+                  replace(' multirow="4" ', ' ').\
+                  replace('valignment="middle"', 'valignment="top"').\
+                  replace(' topline="true" ', ' ')
                 # remove bottom line of previous multirow-part cell
-                if span > 1:
-                  document.body[bottom_index] = document.body[bottom_index].replace(' bottomline="true" ', ' ')
-                # save index to be later able to remove bottom line
-                bottom_index = begin_cell2
-                # remove the multirow tag, set the valignment to top
-                # and remove the top line
-                document.body[begin_cell2] = document.body[begin_cell2].replace(' multirow="4" ', ' ')
-                document.body[begin_cell2] = document.body[begin_cell2].replace('valignment="middle"', 'valignment="top"')
-                document.body[begin_cell2] = document.body[begin_cell2].replace(' topline="true" ', ' ')
-              else:
-                break
-              begin_row2 = find_token(document.body, '</row>', begin_row2)
-            # remove the multirow tag, set the valignment to top,
-            # remove the the bottom line and offset
-            document.body[begin_cell] = document.body[begin_cell].replace(' multirow="3" ', ' ')
-            document.body[begin_cell] = document.body[begin_cell].replace('valignment="middle"', 'valignment="top"')
-            # only remove the bottom line if it is not a single cell as multirow
-            if span > 1:
-              document.body[begin_cell] = document.body[begin_cell].replace(' bottomline="true" ', ' ')
-            document.body[begin_cell] = document.body[begin_cell].replace(' mroffset=', '')
-            # store the offset and remove it
-            begin = document.body[begin_cell].find('"')
-            end = document.body[begin_cell].find('" ', begin)
-            offset = document.body[begin_cell][begin + 1:end]
-            document.body[begin_cell] = document.body[begin_cell].replace(document.body[begin_cell][begin:end + 1], '')
-            # write ERT to create the multirow cell
-            cend = find_token(document.body, "</cell>", begin_cell)
-            if cend == -1:
-              document.warning("Malformed LyX document: Could not find end of tabular cell.")
-              i += 1
-              continue
-            blay = find_token(document.body, "\\begin_layout", begin_cell, cend)
+                lineno = multirows[r-1][col][0]
+                document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ')
+            # revert beginning cell
+            bcell = multirows[row][col][0]
+            ecell = multirows[row][col][1]
+            offset = get_option_value(document.body[bcell], "mroffset")
+            document.body[bcell] = document.body[bcell].\
+              replace(' multirow="3" ', ' ').\
+              replace('valignment="middle"', 'valignment="top"')
+            # remove mroffset option
+            document.body[bcell] = rgx.sub('', document.body[bcell])
+            
+            blay = find_token(document.body, "\\begin_layout", bcell, ecell)
             if blay == -1:
               document.warning("Can't find layout for cell!")
-              i = j
               continue
             bend = find_end_of_layout(document.body, blay)
             if bend == -1:
               document.warning("Can't find end of layout for cell!")
-              i = cend
               continue
             # do the later one first, so as not to mess up the numbering
             # we are wrapping the whole cell in this ert
             # so before the end of the layout...
             document.body[bend:bend] = put_cmd_in_ert("}")
             # ...and after the beginning
-            document.body[blay + 1:blay + 1] = put_cmd_in_ert("\\multirow{" + str(span) + "}{" + col_width + "}[" + offset + "]{")
-          # set index to end of current cell (that is now much bigger due to the ERT addition)
-          begin_cell = find_token(document.body, '</cell>', begin_cell) + 1
-        begin_row = find_token(document.body, '</row>', begin_row) + 1
+            document.body[blay + 1:blay + 1] = \
+              put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}[" \
+                  + offset + "]{")
+
+        # on to the next table
+        begin_table = end_table
+
 
-      # go to the next table
-      begin_table = end_table + 1
+def revert_script(document):
+    " Convert subscript/superscript inset to TeX code "
+    i = 0
+    foundsubscript = False
+    while 1:
+        i = find_token(document.body, '\\begin_inset script', i)
+        if i == -1:
+            break
+        z = find_end_of_inset(document.body, i)
+        if z == -1:
+            document.warning("Malformed LyX document: Can't find end of script inset.")
+            i += 1
+            continue
+        blay = find_token(document.body, "\\begin_layout", i, z)
+        if blay == -1:
+            document.warning("Malformed LyX document: Can't find layout in script inset.")
+            i = z
+            continue
+
+        if check_token(document.body[i], "\\begin_inset script subscript"):
+            subst = '\\textsubscript{'
+            foundsubscript = True
+        elif check_token(document.body[i], "\\begin_inset script superscript"):
+            subst = '\\textsuperscript{'
+        else:
+            document.warning("Malformed LyX document: Unknown type of script inset.")
+            i = z
+            continue
+        bend = find_end_of_layout(document.body, blay)
+        if bend == -1 or bend > z:
+            document.warning("Malformed LyX document: Can't find end of layout in script inset.")
+            i = z
+            continue
+        # remove the \end_layout \end_inset pair
+        document.body[bend:z + 1] = put_cmd_in_ert("}")
+        document.body[i:blay + 1] = put_cmd_in_ert(subst)
+        i += 1
+    # these classes provide a \textsubscript command:
+    # FIXME: Would be nice if we could use the information of the .layout file here
+    classes = ["memoir", "scrartcl", "scrbook", "scrlttr2", "scrreprt"]
+    if foundsubscript and find_token_exact(classes, document.textclass, 0) == -1:
+        add_to_preamble(document, ['\\usepackage{subscript}'])
+
+
+def convert_use_xetex(document):
+    " convert \\use_xetex to \\use_non_tex_fonts "
+    i = 0
+    i = find_token(document.header, "\\use_xetex", 0)
+    if i == -1:
+        return
+    
+    val = get_value(document.header, "\\use_xetex", 0)
+    document.header[i] = "\\use_non_tex_fonts " + val
+
+
+def revert_use_xetex(document):
+    " revert \\use_non_tex_fonts to \\use_xetex "
+    i = 0
+    i = find_token(document.header, "\\use_non_tex_fonts", 0)
+    if i == -1:
+        document.warning("Malformed document. No \\use_non_tex_fonts param!")
+        return
+
+    val = get_value(document.header, "\\use_non_tex_fonts", 0)
+    document.header[i] = "\\use_xetex " + val
+
+
+def revert_labeling(document):
+    koma = ("scrartcl", "scrarticle-beamer", "scrbook", "scrlettr",
+        "scrlttr2", "scrreprt")
+    if document.textclass in koma:
+        return
+    i = 0
+    while True:
+        i = find_token_exact(document.body, "\\begin_layout Labeling", i)
+        if i == -1:
+            return
+        document.body[i] = "\\begin_layout List"
+
+
+def revert_langpack(document):
+    " revert \\language_package parameter "
+    i = 0
+    i = find_token(document.header, "\\language_package", 0)
+    if i == -1:
+        document.warning("Malformed document. No \\language_package param!")
+        return
+
+    del document.header[i]
+
+
+def convert_langpack(document):
+    " Add \\language_package parameter "
+    i = find_token(document.header, "\language" , 0)
+    if i == -1:
+        document.warning("Malformed document. No \\language defined!")
+        return
+
+    document.header.insert(i + 1, "\\language_package default")
+
+
+def revert_tabularwidth(document):
+  i = 0
+  while True:
+    i = find_token(document.body, "\\begin_inset Tabular", i)
+    if i == -1:
+      return
+    j = find_end_of_inset(document.body, i)
+    if j == -1:
+      document.warning("Unable to find end of Tabular inset at line " + str(i))
+      i += 1
+      continue
+    i += 1
+    features = find_token(document.body, "<features", i, j)
+    if features == -1:
+      document.warning("Can't find any features in Tabular inset at line " + str(i))
+      i = j
+      continue
+    if document.body[features].find('alignment="tabularwidth"') != -1:
+      remove_option(document.body, features, 'tabularwidth')
 
-    if multirowOffset == True:
-        add_to_preamble(document, ["\\usepackage{multirow}"])
+def revert_html_css_as_file(document):
+  if not del_token(document.header, '\\html_css_as_file', 0):
+    document.warning("Malformed LyX document: Missing \\html_css_as_file.")
 
 
 ##
@@ -2318,7 +2483,7 @@ convert = [[346, []],
            [368, []],
            [369, [convert_author_id]],
            [370, []],
-           [371, []],
+           [371, [convert_mhchem]],
            [372, []],
            [373, [merge_gbrief]],
            [374, []],
@@ -2354,10 +2519,22 @@ convert = [[346, []],
            [404, [convert_prettyref]],
            [405, []],
            [406, [convert_passthru]],
-           [407, []]
+           [407, []],
+           [408, []],
+           [409, [convert_use_xetex]],
+           [410, []],
+           [411, [convert_langpack]],
+           [412, []],
+           [413, []]
 ]
 
-revert =  [[406, [revert_multirowOffset]],
+revert =  [[412, [revert_html_css_as_file]],
+           [411, [revert_tabularwidth]],
+           [410, [revert_langpack]],
+           [409, [revert_labeling]],
+           [408, [revert_use_xetex]],
+           [407, [revert_script]],
+           [406, [revert_multirowOffset]],
            [405, [revert_passthru]],
            [404, []],
            [403, [revert_refstyle]],