Adds basic import command with document file copy.

Still lot to be improve on ui and features.

papers/commands/__init__.py

@@ -1,5 +1,6 @@
 import add_cmd
 import add_library_cmd
+import import_cmd
 import init_cmd
 import list_cmd
 import open_cmd

papers/commands/import_cmd.py

@@ -0,0 +1,55 @@
+import os
+import sys
+import shutil
+
+from .. import repo
+from ..paper import Paper, NoDocumentFile
+
+
+def parser(subparsers, config):
+    parser = subparsers.add_parser('import',
+            help='import paper(s) to the repository')
+    parser.add_argument('bibpath',
+            help='path to bibtex, bibtexml or bibyaml file (or directory)')
+    parser.add_argument('-c', '--copy', action='store_true', default=None,
+            help="copy document files into library directory (default)")
+    parser.add_argument('-C', '--nocopy', action='store_false', dest='copy',
+            help="don't copy document files (opposite of -c)")
+    return parser
+
+
+def command(config, bibpath, copy):
+    """
+    :param pdffilepath  path (no url yet) to a pdf or ps file
+    :param bibtex       bibtex file (in .bib, .bibml or .yaml format.
+    """
+    if copy is None:
+        copy = config.get('papers', 'import-copy')
+    rp = repo.Repository.from_directory()
+    # Get directory for document
+    doc_path = rp.get_document_directory(config)
+    if not (os.path.exists(doc_path) and os.path.isdir(doc_path)):
+        print "Document directory %s, does not exist." % doc_path
+        sys.exit(1)
+    # Extract papers from bib
+    papers = Paper.many_from_path(bibpath)
+    for p in papers:
+        doc_file = None
+        try:
+            file_path = p.get_document_file_from_bibdata(remove=True)
+            if os.path.exists(file_path):
+                doc_file = file_path
+            else:
+                print "File does not exist for %s." % p.citekey
+        except NoDocumentFile:
+            print "No file for %s." % p.citekey
+        rp.add_paper(p)
+        if doc_file:
+            if copy:
+                ext = os.path.splitext(doc_file)[1]
+                new_doc_file = os.path.join(doc_path, p.citekey + ext)
+                shutil.copy(doc_file, new_doc_file)
+            else:
+                new_doc_file = doc_file
+            p.set_document(new_doc_file)
+            rp.add_or_update(p)

papers/commands/open_cmd.py

@@ -12,7 +12,6 @@ def parser(subparsers, config):
 
 
 def command(config, citekey):
-    print config.get('papers', 'open-cmd')
     rp = repo.Repository.from_directory()
     paper = rp.paper_from_any(citekey, fatal=True)
     try:

papers/configs.py

@@ -0,0 +1,32 @@
+import os
+import ConfigParser
+
+
+DEFAULT_OPEN_CMD = 'open'
+DEFAULT_EDIT_CMD = 'vim'
+
+DEFAULT_IMPORT_COPY = 'yes'
+DEFAULT_IMPORT_MOVE = 'no'
+
+CONFIG = ConfigParser.SafeConfigParser({
+    'open-cmd': DEFAULT_OPEN_CMD,
+    'edit-cmd': DEFAULT_EDIT_CMD,
+    'import-copy': DEFAULT_IMPORT_COPY,
+    'import-move': DEFAULT_IMPORT_MOVE,
+    })
+CONFIG.add_section('papers')
+
+
+def read_config():
+    CONFIG.read(os.path.expanduser('~/.papersrc'))
+    return CONFIG
+
+
+def get_boolean(value, default):
+    value = str(value).lower()
+    if value in ('yes', 'true', 't', 'y', '1'):
+        return True
+    elif value in ('no', 'false', 'f', 'n', '0'):
+        return False
+    else:
+        return 0

papers/files.py

@@ -29,13 +29,18 @@ try:
     EDITOR = os.environ['EDITOR']
 except KeyError:
     EDITOR = 'nano'
+BIB_EXTENSIONS = ['.bib', '.bibyaml', '.bibml', '.yaml']
+
+
+def clean_path(path):
+    return os.path.abspath(os.path.expanduser(path))
 
 
 def find_papersdir():
     """Find .papers directory in this directory and the parent directories"""
     global _papersdir
     if _papersdir is None:
-        curdir = os.path.abspath(os.getcwd())
+        curdir = os.path.abspath('')
         while curdir != '':
             if (os.path.exists(curdir + '/.papers')
                     and os.path.isdir(curdir + '/.papers')):

papers/paper.py

@@ -105,10 +105,10 @@ class Paper(object):
         citekey = u'{}{}'.format(u''.join(first_author.last()), year)
         return str2citekey(citekey)
 
-    def set_pdf(self, pdfpath):
+    def set_document(self, docpath):
-        fullpdfpath = os.path.abspath(pdfpath)
+        fullpdfpath = os.path.abspath(docpath)
         files.check_file(fullpdfpath)
-        name, ext = files.name_from_path(pdfpath)
+        name, ext = files.name_from_path(docpath)
         self.metadata['filename'] = name
         self.metadata['extension'] = ext
         self.metadata['path'] = fullpdfpath
@@ -124,6 +124,30 @@ class Paper(object):
         files.save_bibdata(bibdata, bib_filepath)
         files.save_meta(self.metadata, meta_filepath)
 
+    def get_document_file_from_bibdata(self, remove=False):
+        """Try extracting document file from bib data.
+        Raises NoDocumentFile if not found.
+
+        Parameters:
+        -----------
+        remove: default: False
+            remove field after extracting information
+        """
+        try:
+            field = self.bibentry.fields['file']
+            # Check if this is mendeley specific
+            for f in field.split(':'):
+                if len(f) > 0:
+                    break
+            if remove:
+                self.bibentry.fields.pop('file')
+            # This is a hck for Mendeley. Make clean
+            if f[0] != '/':
+                f = '/' + f
+            return f
+        except (KeyError, IndexError):
+            raise(NoDocumentFile('No file found in bib data.'))
+
     @classmethod
     def load(cls, bibpath, metapath=None):
         key, entry = cls.get_bibentry(bibpath)
@@ -148,7 +172,15 @@ class Paper(object):
         return BASE_META.copy()
 
     @classmethod
-    def many_from_bib(cls, bibpath):
+    def many_from_path(cls, bibpath):
-        bib_data = files.load_externalbibfile(bibpath)
+        """Extract list of papers found in bibliographic files in path.
-        return [Paper(bibentry=bib_data.entries[k], citekey=k)
+        """
-                for k in bib_data.entries]
+        bibpath = files.clean_path(bibpath)
+        if os.path.isdir(bibpath):
+            all_files = [os.path.join(bibpath, f) for f in os.listdir(bibpath)
+                    if os.path.splitext(f)[-1] in files.BIB_EXTENSIONS]
+        else:
+            all_files = [bibpath]
+        bib_data = [files.load_externalbibfile(f) for f in all_files]
+        return [Paper(bibentry=b.entries[k], citekey=k)
+                for b in bib_data for k in b.entries]

papers/papers

@@ -6,24 +6,21 @@ import os
 
 import argparse
 import collections
-import ConfigParser
 
+from papers import configs
 from papers import commands
 
 cmds = collections.OrderedDict([
         ('init', commands.init_cmd),
         ('add', commands.add_cmd),
         ('add_library', commands.add_library_cmd),
+        ('import', commands.import_cmd),
         ('list', commands.list_cmd),
         ('open', commands.open_cmd),
         ('websearch', commands.websearch_cmd)
         ])
 
-config = ConfigParser.SafeConfigParser()
+config = configs.read_config()
-config.add_section('papers')
-config.set('papers', 'open-cmd', 'open')
-config.set('papers', 'edit-cmd', 'vim')
-config.read(os.path.expanduser('~/.papersrc'))
 
 parser = argparse.ArgumentParser(description="research papers repository")
 subparsers = parser.add_subparsers(title="valid commands", dest="command")

papers/repo.py

@@ -9,6 +9,7 @@ ALPHABET = 'abcdefghijklmopqrstuvwxyz'
 BASE_FILE = 'papers.yaml'
 BIB_DIR = 'bibdata'
 META_DIR = 'meta'
+DOC_DIR = 'doc'
 
 
 class Repository(object):
@@ -60,9 +61,9 @@ class Repository(object):
 
     # creating new papers
 
-    def add_paper_from_paths(self, pdfpath, bibpath):
+    def add_paper_from_paths(self, docpath, bibpath):
         p = Paper.load(bibpath)
-        p.set_pdf(pdfpath)
+        p.set_document(docpath)
         self.add_paper(p)
 
     def add_paper(self, p):
@@ -83,7 +84,7 @@ class Repository(object):
         if not paper.citekey in self.citekeys:
             self.add_paper(paper)
         else:
-            paper.save_paper(paper)
+            self.save_paper(paper)
 
     def save_paper(self, paper):
         if not paper.citekey in self.citekeys:
@@ -129,6 +130,12 @@ class Repository(object):
         else:
             raise(ValueError("%s is not a valid paper file." % file_))
 
+    def get_document_directory(self, config):
+        if config.has_option('papers', 'document-directory'):
+            return config.get('papers', 'document-directory')
+        else:
+            return os.path.join(self.papersdir, DOC_DIR)
+
     @classmethod
     def from_directory(cls, papersdir=None):
         repo = cls()