Merge branch 'develop' into texnote

TODO

@@ -1,7 +1,6 @@
 TODO list
 =========
 - manage cross-references
-+ labels
 - find (authors) duplicates
 + remove command
 - stats command

papers/color.py

@@ -24,7 +24,7 @@ filepath = cyan
 def dye(s, color=end, bold=False):
     assert color[0] == '\033'
     if bold:
-        s = '\033[1' + s[3:]
+        color = '\033[1' + color[3:]
     return color + s + end
 
 _dye = dye

papers/commands/__init__.py

@@ -8,5 +8,6 @@ import open_cmd
 import edit_cmd
 import remove_cmd
 import websearch_cmd
-import tags_cmd
+import tag_cmd
 import attach_cmd
+import update_cmd

papers/commands/add_cmd.py

@@ -10,7 +10,7 @@ def parser(subparsers, config):
     parser.add_argument('-b', '--bibfile',
                         help='bibtex, bibtexml or bibyaml file', default=None)
     parser.add_argument('-d', '--docfile', help='pdf or ps file', default=None)
-    parser.add_argument('-l', '--label', help='label associated to the paper',
+    parser.add_argument('-t', '--tags', help='tags associated to the paper, separated by commas',
                         default=None)
     parser.add_argument('-c', '--copy', action='store_true', default=None,
             help="copy document files into library directory (default)")
@@ -19,7 +19,7 @@ def parser(subparsers, config):
     return parser
 
 
-def command(config, ui, bibfile, docfile, label, copy):
+def command(config, ui, bibfile, docfile, tags, copy):
     """
     :param bibfile: bibtex file (in .bib, .bibml or .yaml format.
     :param docfile: path (no url yet) to a pdf or ps file
@@ -37,15 +37,15 @@ def command(config, ui, bibfile, docfile, label, copy):
                 cont = False
             except Exception:
                 cont = ui.input_yn(
-                    question='Invalid bibfile. Edit again or abort?',
+                    question='Invalid bibfile. Edit again ?',
                     default='y')
                 if not cont:
                     ui.exit()
         p = Paper(bibentry=bib, citekey=key)
     else:
         p = Paper.load(bibfile)
-    if label is not None:
+    if tags is not None:
-        p.metadata['labels'] = label.split()
+        p.tags = set(tags.split(','))
     # Check if another doc file is specified in bibtex
     docfile2 = extract_doc_path_from_bibdata(p, ui)
     if docfile is None:

papers/commands/helpers.py

@@ -1,5 +1,6 @@
 from .. import files
 from .. import color
+from .. import pretty
 from ..repo import InvalidReference
 from ..paper import NoDocumentFile
 
@@ -51,3 +52,16 @@ def parse_reference(ui, rp, ref):
 def parse_references(ui, rp, refs):
     citekeys = [parse_reference(ui, rp, ref) for ref in refs]
     return citekeys
+
+def paper_oneliner(p, n = 0, citekey_only = False):
+    if citekey_only:
+        return p.citekey
+    else:
+        bibdesc = pretty.bib_oneliner(p.bibentry)
+        return (u'{num:d}: [{citekey}] {descr} {tags}'.format(
+            num=int(n),
+            citekey=color.dye(p.citekey, color.purple),
+            descr=bibdesc,
+            tags=color.dye(' '.join(p.tags),
+                           color.purple, bold=True),
+            )).encode('utf-8')

papers/commands/init_cmd.py

@@ -24,13 +24,14 @@ def command(config, ui, path, doc_dir):
     else:
         papersdir = os.path.join(os.getcwd(), path)
         configs.add_and_write_option('papers', 'papers-directory', papersdir)
-    if not os.path.exists(papersdir):
+    if os.path.exists(papersdir):
-        ui.print_('Initializing papers in {}.'.format(
+        if len(os.listdir(papersdir)) > 0:
-                  color.dye(papersdir, color.filepath)))
+            ui.error('directory {} is not empty.'.format(
-        repo = Repository()
+                                 color.dye(papersdir, color.filepath)))
-        repo.init(papersdir)  # Creates directories
+            ui.exit()
-        repo.save()  # Saves empty repository description
+
-    else:
+    ui.print_('Initializing papers in {}.'.format(
-        ui.error('papers already present in {}.'.format(
+              color.dye(papersdir, color.filepath)))
-                 color.dye(papersdir, color.filepath)))
+    repo = Repository()
-        ui.exit()
+    repo.init(papersdir)  # Creates directories
+    repo.save()  # Saves empty repository description

papers/commands/list_cmd.py

@@ -1,6 +1,7 @@
 from .. import pretty
 from .. import repo
 from .. import color
+from . import helpers
 
 
 def parser(subparsers, config):
@@ -9,7 +10,7 @@ def parser(subparsers, config):
             default=False, dest='citekeys',
             help='Only returns citekeys of matching papers.')
     parser.add_argument('query', nargs='*',
-            help='Paper query (e.g. "year: 2000" or "labels: math")')
+            help='Paper query (e.g. "year: 2000" or "tags: math")')
     return parser
 
 
@@ -17,20 +18,7 @@ def command(config, ui, citekeys, query):
     rp = repo.Repository.from_directory(config)
     papers = [(n, p) for n, p in enumerate(rp.all_papers())
               if test_paper(query, p)]
-    if citekeys:
+    ui.print_('\n'.join(helpers.paper_oneliner(p, n = n, citekey_only = citekeys) for n, p in papers))
-        paper_strings = [p.citekey for n, p in papers]
-    else:
-        paper_strings = []
-        for n, p in papers:
-            bibdesc = pretty.bib_oneliner(p.bibentry)
-            paper_strings.append((u'{num:d}: [{citekey}] {descr} {labels}'.format(
-                num=int(n),
-                citekey=color.dye(p.citekey, color.purple),
-                descr=bibdesc,
-                labels=color.dye(' '.join(p.metadata.get('labels', [])),
-                                 color.purple, bold=True),
-                )).encode('utf-8'))
-    ui.print_('\n'.join(paper_strings))
 
 
 # TODO author is not implemented, should we do it by last name only or more
@@ -45,8 +33,8 @@ def test_paper(query_string, p):
         field = tmp[0]
         value = tmp[1]
 
-        if field in ['labels', 'l', 'tags', 't']:
+        if field in ['tags', 't']:
-            if value not in p.metadata['labels']:
+            if value not in p.tags:
                 return False
         elif field in ['author', 'authors', 'a']:  # that is the very ugly
             if not 'author' in p.bibentry.persons:

papers/commands/tag_cmd.py

@@ -0,0 +1,59 @@
+"""
+This command is all about tags.
+The different use cases are :
+1. > papers tag
+    Returns the list of all tags
+2. > papers tag citekey
+    Return the list of tags of the given citekey
+3. > papers tag citekey math
+    Add 'math' to the list of tags of the given citekey
+4. > papers tag citekey :math
+    Remove 'math' for the list of tags of the given citekey
+5. > papers tag citekey math,romance,:war
+    Add 'math' and 'romance' tags to the given citekey, and remove the 'war' tag
+6. > papers tag math
+    If 'math' is not a citekey, then display all papers with the tag 'math'
+"""
+
+from ..repo import Repository, InvalidReference
+from . import helpers
+
+def parser(subparsers, config):
+    parser = subparsers.add_parser('tag', help="add, remove and show tags")
+    parser.add_argument('referenceOrTag', nargs='?', default = None,
+                        help='reference to the paper (citekey or number), or '
+                             'tag.')
+    parser.add_argument('tags', nargs='?', default = None,
+                        help='If the previous argument was a reference, then '
+                             'then a list of tags separated by commas.')
+    # TODO find a way to display clear help for multiple command semantics,
+    #      indistinguisable for argparse. (fabien, 201306)
+    return parser
+
+def command(config, ui, referenceOrTag, tags):
+    """Add, remove and show tags"""
+    rp = Repository.from_directory(config)
+
+    if referenceOrTag is None:
+        for tag in rp.get_tags():
+            ui.print_(tag)
+    else:
+        try:
+            citekey = rp.citekey_from_ref(referenceOrTag)
+            p = rp.get_paper(citekey)
+            if tags is None:
+                ui.print_(' '.join(p.tags))
+            else:
+                tags = tags.split(',')
+                for tag in tags:
+                    if tag[0] == ':':
+                        p.remove_tag(tag[1:])
+                    else:
+                        p.add_tag(tag)
+                rp.save_paper(p)
+        except InvalidReference:
+            tag = referenceOrTag
+            papers_list = [(p, n) for n, p in enumerate(rp.all_papers())
+                           if tag in p.tags]
+            ui.print_('\n'.join(helpers.paper_oneliner(p, n)
+                                for p, n in papers_list))

papers/commands/tags_cmd.py

@@ -1,13 +0,0 @@
-from ..repo import Repository
-
-
-def parser(subparsers, config):
-    parser = subparsers.add_parser('tags', help="list existing tags")
-    return parser
-
-
-def command(config, ui):
-    """List existing tags"""
-    rp = Repository.from_directory(config)
-    for tag in rp.get_labels():
-        ui.print_(tag)

papers/commands/update_cmd.py

@@ -0,0 +1,19 @@
+from .. import repo
+from .. import color
+
+def parser(subparsers, config):
+    parser = subparsers.add_parser('update', help='update the repository to the lastest format')
+    return parser
+
+
+def command(config, ui):
+    rp = repo.Repository.from_directory(config)
+    msg = ("You should backup the paper directory {} before continuing."
+           "Continue ?").format(color.dye(rp.papersdir, color.filepath))
+    sure = ui.input_yn(question=msg, default='n')
+    if sure:
+        for p in rp.all_papers():
+            tags = set(p.metadata['tags'])
+            tags = tags.union(p.metadata['labels'])
+            p.metadata.pop('labels', [])
+            rp.save_paper(p)

papers/commands/websearch_cmd.py

@@ -5,12 +5,13 @@ import urllib
 def parser(subparsers, config):
     parser = subparsers.add_parser('websearch',
                                    help="launch a search on Google Scholar")
-    parser.add_argument("search_string",
+    parser.add_argument("search_string", nargs = '*',
                         help="the search query (anything googly is possible)")
     return parser
 
 
 def command(config, ui, search_string):
+    print search_string
     url = ("https://scholar.google.fr/scholar?q=%s&lr="
-           % (urllib.quote_plus(search_string)))
+           % (urllib.quote_plus(' '.join(search_string))))
     webbrowser.open(url)

papers/files.py

@@ -1,6 +1,7 @@
 import os
 import subprocess
 import tempfile
+from cStringIO import StringIO
 
 import yaml
 
@@ -21,10 +22,10 @@ try:
     import pybtex.database.output.bibyaml
 
 except ImportError:
-    print(ui.dye('error', ui.error) +
+    print(color.dye('error', color.error) +
-          ": you need to install Pybtex; try running 'pip install"
+          ": you need to install Pybtex; try running 'pip install "
           "pybtex' or 'easy_install pybtex'")
-
+    exit(-1)
 
 _papersdir = None
 
@@ -122,7 +123,7 @@ def load_externalbibfile(fullbibpath):
     filename, ext = os.path.splitext(os.path.split(fullbibpath)[1])
     if ext[1:] in FORMATS_INPUT.keys():
         with open(fullbibpath) as f:
-            return parse_bibdata(f, ext[1:])
+            return _parse_bibdata_formated_stream(f, ext[1:])
     else:
         print('{}: {} not recognized format for bibliography'.format(
               color.dye('error', color.error),
@@ -130,14 +131,38 @@ def load_externalbibfile(fullbibpath):
         exit(-1)
 
 
-def parse_bibdata(content, format_):
+def _parse_bibdata_formated_stream(stream, fmt):
-    """Parse bib data from string.
+    """Parse a stream for bibdata, using the supplied format."""
-
+    try:
+        parser = FORMATS_INPUT[fmt].Parser()
+        data = parser.parse_stream(stream)
+        if data.entries.keys() > 0:
+            return data
+    except Exception:
+        pass
+    raise ValueError, 'content format is not recognized.'
+
+def parse_bibdata(content, format_ = None):
+    """Parse bib data from string or stream.
+
+    Raise ValueError if no bibdata is present.
     :content: stream
-    :param format_: (bib|xml|yml)
+    :param format_: (bib|xml|yml) if format is None, tries to recognize the
+                    format automatically.
     """
-    parser = FORMATS_INPUT[format_].Parser()
+    fmts = [format_]
-    return parser.parse_stream(content)
+    if format_ is None:
+        fmts = FORMATS_INPUT.keys()
+        # we need to reuse the content
+        content = content if type(content) == str else str(content.read())
+
+    for fmt in fmts:
+        try:
+            return _parse_bibdata_formated_stream(StringIO(content), fmt)
+        except Exception:
+            pass
+
+    raise ValueError, 'content format is not recognized.'
 
 
 def editor_input(config, initial="", suffix=None):

papers/paper.py

@@ -20,7 +20,7 @@ CITEKEY_EXCLUDE_RE = re.compile('[%s]'
 
 BASE_META = {
     'external-document': None,
-    'labels': [],
+    'tags': [],
     'notes': [],
     }
 
@@ -44,7 +44,7 @@ def get_bibentry_from_file(bibfile):
 def get_bibentry_from_string(content):
     """Extract first entry (supposed to be the only one) from given file.
     """
-    bib_data = files.parse_bibdata(StringIO(content), 'yml')
+    bib_data = files.parse_bibdata(StringIO(content))
     first_key = bib_data.entries.keys()[0]
     first_entry = bib_data.entries[first_key]
     return first_key, first_entry
@@ -69,6 +69,7 @@ def get_safe_metadata(meta):
     base_meta = Paper.create_meta()
     if meta is not None:
         base_meta.update(meta)
+    base_meta['tags'] = set(base_meta['tags'])
     return base_meta
 
 
@@ -250,7 +251,27 @@ class Paper(object):
             return papers
 
 
-class PaperInRepo(Paper):
+    # tags
+
+    @property
+    def tags(self):
+        return self.metadata.setdefault('tags', set())
+
+    @tags.setter
+    def tags(self, value):
+        if not hasattr(value, '__iter__'):
+            raise ValueError, 'tags must be iterables'
+        self.metadata['tags'] = set(value)
+
+    def add_tag(self, tag):
+        self.tags.add(tag)
+
+    def remove_tag(self, tag):
+        """Remove a tag from a paper if present."""
+        self.tags.discard(tag)
+
+
+class PaperInRepo(Paper): # TODO document why this class exists (fabien, 2013/06)
 
     def __init__(self, repo, *args, **kwargs):
         Paper.__init__(self, *args, **kwargs)

papers/papers

@@ -11,18 +11,19 @@ from papers import commands
 from papers import plugins
 
 cmds = collections.OrderedDict([
-        ('init', commands.init_cmd),
+        ('init',        commands.init_cmd),
-        ('add', commands.add_cmd),
+        ('add',         commands.add_cmd),
         ('add_library', commands.add_library_cmd),
-        ('import', commands.import_cmd),
+        ('import',      commands.import_cmd),
-        ('export', commands.export_cmd),
+        ('export',      commands.export_cmd),
-        ('list', commands.list_cmd),
+        ('list',        commands.list_cmd),
-        ('edit', commands.edit_cmd),
+        ('edit',        commands.edit_cmd),
-        ('remove', commands.remove_cmd),
+        ('remove',      commands.remove_cmd),
-        ('open', commands.open_cmd),
+        ('open',        commands.open_cmd),
-        ('websearch', commands.websearch_cmd),
+        ('websearch',   commands.websearch_cmd),
-        ('tags', commands.tags_cmd),
+        ('tag',         commands.tag_cmd),
-        ('attach', commands.attach_cmd),
+        ('attach',      commands.attach_cmd),
+        ('update',      commands.update_cmd),
         ])
 
 config = configs.read_config()

papers/repo.py

@@ -198,11 +198,11 @@ class Repository(object):
             new_doc_file = os.path.join(doc_path, citekey + ext)
             shutil.copy(doc_file, new_doc_file)
 
-    def get_labels(self):
+    def get_tags(self):
-        labels = set()
+        tags = set()
         for p in self.all_papers():
-            labels = labels.union(p.metadata.get('labels', []))
+            tags = tags.union(p.tags)
-        return labels
+        return tags
 
     @classmethod
     def from_directory(cls, config, papersdir=None):

tests/test_paper.py

@@ -24,7 +24,7 @@ entries:
 META = """
 external-document: null
 notes: []
-labels: []
+tags: []
 """