From 4a0cdafd28b4288b8f9fae9db84b0ebd8dfba8a8 Mon Sep 17 00:00:00 2001
From: Robin Krahl <robin.krahl@ireas.org>
Date: Thu, 8 Mar 2018 21:07:16 +0100
Subject: Add basic import implementation

---
 bibtool/cli.py     | 42 ++++++++++++++++++++++++++++++++++++++++--
 bibtool/extract.py | 37 +++++++++++++++++++++++++++++++++++++
 requirements.txt   |  2 ++
 3 files changed, 79 insertions(+), 2 deletions(-)
 create mode 100644 bibtool/extract.py

diff --git a/bibtool/cli.py b/bibtool/cli.py
index 23ceaa4..3356d16 100644
--- a/bibtool/cli.py
+++ b/bibtool/cli.py
@@ -1,6 +1,12 @@
 # Copyright (C) 2018 Robin Krahl <robin.krahl@ireas.org>
 # SPDX-License-Identifier: MIT
 
+import os
+import os.path
+import shutil
+
+import bibtexparser
+import bibtool.extract
 import click
 
 
@@ -15,6 +21,38 @@ def cli():
 
 @cli.command('import')
 @click.argument('filename', type=TYPE_FILE)
-@click.option('--directory', type=TYPE_DIR)
+@click.option('--directory', type=TYPE_DIR, default=os.getcwd())
 def _import(filename, directory):
-    click.echo(filename)
+    bibtex_data = bibtool.extract.get_bibtex_data(filename)
+
+    if not bibtex_data.entries:
+        raise Exception('Did not find any Bibtex entry.')
+    if len(bibtex_data.entries) > 1:
+        raise Exception('Found more than one Bibtex entries.')
+
+    entry = bibtex_data.entries[0]
+    click.echo('Found one Bibtex entry: ' + entry['ID'])
+    if click.confirm('Do you want to edit the entry?'):
+        edited_data = click.edit(bibtexparser.dumps(bibtex_data))
+        if edited_data:
+            bibtex_data = bibtexparser.loads(edited_data)
+            if len(bibtex_data.entries) != 1:
+                raise Exception('The edited data must contain exactly one Bibtex entry.')
+            entry = bibtex_data.entries[0]
+
+    click.confirm('Add {} to the repository?'.format(entry['ID']),
+                  default=True, abort=True)
+
+    bibfilename = os.path.join(directory, entry['ID'] + '.bib')
+    outfileext = os.path.splitext(filename)[1]
+    outfilename = os.path.join(directory, entry['ID'] + outfileext)
+    
+    if os.path.exists(bibfilename) or os.path.exists(outfilename):
+        click.confirm('There is already a document with this ID in the '
+                      'repository. Continue anyway?', abort=True)
+
+    with open(bibfilename, 'w') as f:
+        bibtexparser.dump(bibtex_data, f)
+    shutil.copy(filename, outfilename)
+
+    click.echo('Added {} to the repository.'.format(entry['ID']))
diff --git a/bibtool/extract.py b/bibtool/extract.py
new file mode 100644
index 0000000..6409f1b
--- /dev/null
+++ b/bibtool/extract.py
@@ -0,0 +1,37 @@
+# Copyright (C) 2018 Robin Krahl <robin.krahl@ireas.org>
+# SPDX-License-Identifier: MIT
+
+import mimetypes
+import urllib.request
+
+import bibtexparser
+import PyPDF2.pdf
+
+
+PDF_INFO_DOI = '/doi'
+
+
+def _get_data_for_doi(doi):
+    url = 'http://dx.doi.org/{}'.format(doi)
+    headers = {'Accept': 'text/bibliography; style=bibtex'}
+    request = urllib.request.Request(url, headers=headers)
+    with urllib.request.urlopen(request) as response:
+        bibtex = response.read().decode('utf-8')
+    return bibtexparser.loads(bibtex)
+
+
+def _handle_pdf(filename):
+    with open(filename, 'rb') as f:
+        reader = PyPDF2.pdf.PdfFileReader(f)
+        if PDF_INFO_DOI not in reader.documentInfo:
+            raise Exception('PDF file does not have doi header')
+        doi = reader.documentInfo[PDF_INFO_DOI]
+
+    return _get_data_for_doi(doi)
+
+
+def get_bibtex_data(filename):
+    mtype, _ = mimetypes.guess_type(filename)
+    if mtype == 'application/pdf':
+        return _handle_pdf(filename)
+    raise Exception('Unsupported file type ' + str(mtype))
diff --git a/requirements.txt b/requirements.txt
index dca9a90..620d787 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1 +1,3 @@
+bibtexparser
 click
+PyPDF2
-- 
cgit v1.2.3