Add raw text option to text conversion (lesmianator will need it).

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index 94441db..5f6f0a2 100644 (file)
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -19,25 +19,25 @@ from Texml.processor import process
 from lxml import etree
 from lxml.etree import XMLSyntaxError, XSLTApplyError
 
 from lxml import etree
 from lxml.etree import XMLSyntaxError, XSLTApplyError
 

+from librarian.dcparser import Person

 from librarian.parser import WLDocument
 from librarian.parser import WLDocument

-from librarian import ParseError
+from librarian import ParseError, DCNS, get_resource

 from librarian import functions
 
 
 from librarian import functions
 
 

-

 functions.reg_substitute_entities()
 functions.reg_substitute_entities()

-functions.reg_person_name()

 functions.reg_strip()
 functions.reg_starts_white()
 functions.reg_ends_white()
 functions.reg_strip()
 functions.reg_starts_white()
 functions.reg_ends_white()

+functions.reg_texcommand()

 
 STYLESHEETS = {
 
 STYLESHEETS = {

-    'wl2tex': 'xslt/wl2tex.xslt',
+    'wl2tex': 'pdf/wl2tex.xslt',

 }
 
 
 }
 
 

-def insert_tags(doc, split_re, tagname):
-    """ inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree 
+def insert_tags(doc, split_re, tagname, exclude=None):
+    """ inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree

 
     >>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>');
     >>> insert_tags(t, re.compile('-'), 'd');
 
     >>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>');
     >>> insert_tags(t, re.compile('-'), 'd');


@@ -45,46 +45,47 @@ def insert_tags(doc, split_re, tagname):
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
 
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
 

-    for elem in doc.iter():
-        try:
-            if elem.text:
-                chunks = split_re.split(elem.text)
-                while len(chunks) > 1:
-                    ins = etree.Element(tagname)
-                    ins.tail = chunks.pop()
-                    elem.insert(0, ins)
-                elem.text = chunks.pop(0)
-            if elem.tail:
-                chunks = split_re.split(elem.tail)
-                parent = elem.getparent()
-                ins_index = parent.index(elem) + 1
-                while len(chunks) > 1:
-                    ins = etree.Element(tagname)
-                    ins.tail = chunks.pop()
-                    parent.insert(ins_index, ins)
-                elem.tail = chunks.pop(0)
-        except TypeError, e:
-            # element with no children, like comment
-            pass
+    for elem in doc.iter(tag=etree.Element):
+        if exclude and elem.tag in exclude:
+            continue
+        if elem.text:
+            chunks = split_re.split(elem.text)
+            while len(chunks) > 1:
+                ins = etree.Element(tagname)
+                ins.tail = chunks.pop()
+                elem.insert(0, ins)
+            elem.text = chunks.pop(0)
+        if elem.tail:
+            chunks = split_re.split(elem.tail)
+            parent = elem.getparent()
+            ins_index = parent.index(elem) + 1
+            while len(chunks) > 1:
+                ins = etree.Element(tagname)
+                ins.tail = chunks.pop()
+                parent.insert(ins_index, ins)
+            elem.tail = chunks.pop(0)

 
 
 def substitute_hyphens(doc):
 
 
 def substitute_hyphens(doc):

-    insert_tags(doc, 
+    insert_tags(doc,

                 re.compile("(?<=[^-\s])-(?=[^-\s])"),
                 re.compile("(?<=[^-\s])-(?=[^-\s])"),

-                "dywiz")
+                "dywiz",
+                exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+                )

 
 
 def fix_hanging(doc):
 
 
 def fix_hanging(doc):

-    insert_tags(doc, 
+    insert_tags(doc,

                 re.compile("(?<=\s\w)\s+"),
                 re.compile("(?<=\s\w)\s+"),

-                "nbsp")
+                "nbsp",
+                exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+                )

 
 
 def move_motifs_inside(doc):
     """ moves motifs to be into block elements """
     for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|//dramat_wierszowany_l|//dramat_wierszowany_lp|//dramat_wspolczesny'):
         for motif in master.xpath('motyw'):
 
 
 def move_motifs_inside(doc):
     """ moves motifs to be into block elements """
     for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|//dramat_wierszowany_l|//dramat_wierszowany_lp|//dramat_wspolczesny'):
         for motif in master.xpath('motyw'):

-            print motif.text

             for sib in motif.itersiblings():
                 if sib.tag not in ('sekcja_swiatlo', 'sekcja_asterysk', 'separator_linia', 'begin', 'end', 'motyw', 'extra', 'uwaga'):
                     # motif shouldn't have a tail - it would be untagged text
             for sib in motif.itersiblings():
                 if sib.tag not in ('sekcja_swiatlo', 'sekcja_asterysk', 'separator_linia', 'begin', 'end', 'motyw', 'extra', 'uwaga'):
                     # motif shouldn't have a tail - it would be untagged text


@@ -126,15 +127,27 @@ def hack_motifs(doc):
                 break
 
 
                 break
 
 

-def get_resource(path):
-    return os.path.join(os.path.dirname(__file__), path)
+def parse_creator(doc):
+    """ find all dc:creator and dc.contributor tags and add *_parsed versions with forenames first """
+    for person in doc.xpath("|".join('//dc:'+(tag) for tag in (
+                    'creator', 'contributor.translator', 'contributor.editor', 'contributor.technical_editor')),
+                    namespaces = {'dc': str(DCNS)})[::-1]:
+        if not person.text:
+            continue
+        p = Person.from_text(person.text)
+        person_parsed = deepcopy(person)
+        person_parsed.tag = person.tag + '_parsed'
+        person_parsed.set('sortkey', person.text)
+        person_parsed.text = p.readable()
+        person.getparent().insert(0, person_parsed)
+

 
 def get_stylesheet(name):
     return get_resource(STYLESHEETS[name])
 
 
 def package_available(package, args='', verbose=False):
 
 def get_stylesheet(name):
     return get_resource(STYLESHEETS[name])
 
 
 def package_available(package, args='', verbose=False):

-    """ check if a verion of a latex package accepting given args is available """  
+    """ check if a verion of a latex package accepting given args is available """

     tempdir = mkdtemp('-wl2pdf-test')
     fpath = os.path.join(tempdir, 'test.tex')
     f = open(fpath, 'w')
     tempdir = mkdtemp('-wl2pdf-test')
     fpath = os.path.join(tempdir, 'test.tex')
     f = open(fpath, 'w')


@@ -153,34 +166,42 @@ def package_available(package, args='', verbose=False):
     return p == 0
 
 
     return p == 0
 
 

-def transform(provider, slug, output_file=None, output_dir=None, make_dir=False, verbose=False, save_tex=None):
+def transform(provider, slug=None, file_path=None,
+              output_file=None, output_dir=None, make_dir=False, verbose=False, save_tex=None, morefloats=None):

     """ produces a PDF file with XeLaTeX
 
     provider: a DocProvider
     slug: slug of file to process, available by provider
     """ produces a PDF file with XeLaTeX
 
     provider: a DocProvider
     slug: slug of file to process, available by provider

+    file_path can be provided instead of a slug

     output_file: file-like object or path to output file
     output_dir: path to directory to save output file to; either this or output_file must be present
     make_dir: writes output to <output_dir>/<author>/<slug>.pdf istead of <output_dir>/<slug>.pdf
     verbose: prints all output from LaTeX
     save_tex: path to save the intermediary LaTeX file to
     output_file: file-like object or path to output file
     output_dir: path to directory to save output file to; either this or output_file must be present
     make_dir: writes output to <output_dir>/<author>/<slug>.pdf istead of <output_dir>/<slug>.pdf
     verbose: prints all output from LaTeX
     save_tex: path to save the intermediary LaTeX file to

+    morefloats (old/new/none): force specific morefloats

     """
 
     # Parse XSLT
     try:
     """
 
     # Parse XSLT
     try:

-        # check for latex packages
-        if not package_available('morefloats', 'maxfloats=19', verbose=verbose):
-            document.edoc.getroot().set('old-morefloats', 'yes')
-            print >> sys.stderr, """
-==============================================================================
-LaTeX `morefloats' package is older than v.1.0c or not available at all.
-Some documents with many motifs in long stanzas or paragraphs may not compile.
-=============================================================================="""
+        if file_path:
+            if slug:
+                raise ValueError('slug or file_path should be specified, not both')
+            document = load_including_children(provider, file_path=file_path)
+        else:
+            if not slug:
+                raise ValueError('either slug or file_path should be specified')
+            document = load_including_children(provider, slug=slug)

 
 

-        document = load_including_children(provider, slug)
+        # check for LaTeX packages
+        if morefloats:
+            document.edoc.getroot().set('morefloats', morefloats.lower())
+        elif package_available('morefloats', 'maxfloats=19'):
+            document.edoc.getroot().set('morefloats', 'new')

 
         # hack the tree
         move_motifs_inside(document.edoc)
         hack_motifs(document.edoc)
 
         # hack the tree
         move_motifs_inside(document.edoc)
         hack_motifs(document.edoc)

+        parse_creator(document.edoc)

         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)
 
         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)
 


@@ -208,7 +229,7 @@ Some documents with many motifs in long stanzas or paragraphs may not compile.
 
         # LaTeX -> PDF
         shutil.copy(get_resource('pdf/wl.sty'), temp)
 
         # LaTeX -> PDF
         shutil.copy(get_resource('pdf/wl.sty'), temp)

-        shutil.copy(get_resource('pdf/wl-logo.png'), temp)
+        shutil.copy(get_resource('res/wl-logo.png'), temp)

 
         cwd = os.getcwd()
         os.chdir(temp)
 
         cwd = os.getcwd()
         os.chdir(temp)


@@ -229,7 +250,10 @@ Some documents with many motifs in long stanzas or paragraphs may not compile.
                 os.makedirs(output_dir)
             except OSError:
                 pass
                 os.makedirs(output_dir)
             except OSError:
                 pass

-            output_path = os.path.join(output_dir, '%s.pdf' % slug)
+            if slug:
+                output_path = os.path.join(output_dir, '%s.pdf' % slug)
+            else:
+                output_path = os.path.join(output_dir, os.path.splitext(os.path.basename(file_path))[0] + '.pdf')

             shutil.move(pdf_path, output_path)
         else:
             if hasattr(output_file, 'write'):
             shutil.move(pdf_path, output_path)
         else:
             if hasattr(output_file, 'write'):


@@ -246,8 +270,8 @@ Some documents with many motifs in long stanzas or paragraphs may not compile.
         raise ParseError(e)
 
 
         raise ParseError(e)
 
 

-def load_including_children(provider, slug=None, uri=None):
-    """ makes one big xml file with children inserted at end 
+def load_including_children(provider, slug=None, uri=None, file_path=None):
+    """ makes one big xml file with children inserted at end

     either slug or uri must be provided
     """
 
     either slug or uri must be provided
     """
 


@@ -255,15 +279,20 @@ def load_including_children(provider, slug=None, uri=None):
         f = provider.by_uri(uri)
     elif slug:
         f = provider[slug]
         f = provider.by_uri(uri)
     elif slug:
         f = provider[slug]

+    elif file_path:
+        f = open(file_path, 'r')

     else:
     else:

-        raise ValueError('Neither slug nor URI provided for a book.')
+        raise ValueError('Neither slug, URI nor file path provided for a book.')

 
 

-    document = WLDocument.from_file(f, True,
-        parse_dublincore=True,
-        preserve_lines=False)
+    text = f.read().decode('utf-8')
+    text = re.sub(ur"([\u0400-\u04ff]+)", ur"<alien>\1</alien>", text)

 
 

+    document = WLDocument.from_string(text, True,
+        parse_dublincore=True)
+
+    f.close()

     for child_uri in document.book_info.parts:
     for child_uri in document.book_info.parts:

+        print child_uri

         child = load_including_children(provider, uri=child_uri)
         document.edoc.getroot().append(child.edoc.getroot())
         child = load_including_children(provider, uri=child_uri)
         document.edoc.getroot().append(child.edoc.getroot())

-

     return document
     return document