diff --git a/changelog.md b/changelog.md
index 1a3267b..314ccbd 100644
--- a/changelog.md
+++ b/changelog.md
@@ -29,6 +29,8 @@
 
 ### Fixed bugs
 
+- [[#149]](https://github.com/pubs/pubs/issues/149) More robust handling of parsing and citekey errors [(#87)](https://github.com/pubs/pubs/pull/87)
+
 - [[#148]](https://github.com/pubs/pubs/issues/148) Fix compatibility with Pyfakefs 3.7 [(#151)](https://github.com/pubs/pubs/pull/151)
 
 - [[#95]](https://github.com/pubs/pubs/issues/95) Error message when editor is missing [(#141)](https://github.com/pubs/pubs/issues/141)
diff --git a/pubs/commands/add_cmd.py b/pubs/commands/add_cmd.py
index a22c0a5..d193b5e 100644
--- a/pubs/commands/add_cmd.py
+++ b/pubs/commands/add_cmd.py
@@ -11,6 +11,7 @@ from .. import templates
 from .. import apis
 from .. import pretty
 from .. import utils
+from .. import endecoder
 from ..completion import CommaSeparatedTagsCompletion
 
 
@@ -57,12 +58,13 @@ def bibentry_from_editor(conf, ui, rp):
                 bibstruct.verify_bibdata(bibentry)
                 # REFACTOR Generate citykey
                 again = False
-        except ValueError:
+
+        except endecoder.EnDecoder.BibDecodingError:
             again = ui.input_yn(
-                question='Invalid bibfile. Edit again ?',
+                question='Invalid bibfile. Edit again?',
                 default='y')
             if not again:
-                ui.exit(0)
+                ui.exit()
 
     return bibentry
 
diff --git a/pubs/commands/edit_cmd.py b/pubs/commands/edit_cmd.py
index b09c885..8cc175b 100644
--- a/pubs/commands/edit_cmd.py
+++ b/pubs/commands/edit_cmd.py
@@ -64,9 +64,14 @@ def command(conf, args):
                              'as `{}`.'.format(citekey, new_paper.citekey)))
                 else:
                     ui.info(('Paper `{}` was successfully edited.'.format(
-                                                                      citekey)))
+                        citekey)))
             break
 
+        except coder.BibDecodingError:
+            if not ui.input_yn(question="Error parsing bibdata. Edit again?"):
+                ui.error("Aborting, paper not updated.")
+                ui.exit()
+
         except repo.CiteKeyCollision:
             options = ['overwrite', 'edit again', 'abort']
             choice = options[ui.input_choice(
diff --git a/pubs/commands/import_cmd.py b/pubs/commands/import_cmd.py
index 9ccb935..fa6236e 100644
--- a/pubs/commands/import_cmd.py
+++ b/pubs/commands/import_cmd.py
@@ -13,6 +13,10 @@ from ..uis import get_ui
 from ..content import system_path, read_text_file
 
 
+_ABORT_USE_IGNORE_MSG = "Aborting import. Use --ignore-malformed to ignore."
+_IGNORING_MSG = " Ignoring it."
+
+
 def parser(subparsers, conf):
     parser = subparsers.add_parser('import',
             help='import paper(s) to the repository')
@@ -24,10 +28,12 @@ def parser(subparsers, conf):
             help="one or several keys to import from the file")
     parser.add_argument('-O', '--overwrite', action='store_true', default=False,
             help="Overwrite keys already in the database")
+    parser.add_argument('-i', '--ignore-malformed', action='store_true', default=False,
+            help="Ignore malformed and unreadable files and entries")
     return parser
 
 
-def many_from_path(bibpath):
+def many_from_path(ui, bibpath, ignore=False):
     """Extract list of papers found in bibliographic files in path.
 
     The behavior is to:
@@ -49,16 +55,31 @@ def many_from_path(bibpath):
 
     biblist = []
     for filepath in all_files:
-        biblist.append(coder.decode_bibdata(read_text_file(filepath)))
+        try:
+            biblist.append(coder.decode_bibdata(read_text_file(filepath)))
+        except coder.BibDecodingError:
+            error = "Could not parse bibtex at {}.".format(filepath)
+            if ignore:
+                ui.warning(error + _IGNORING_MSG)
+            else:
+                ui.error(error + _ABORT_USE_IGNORE_MSG)
+                ui.exit()
 
     papers = {}
     for b in biblist:
         for k, b in b.items():
+            if k in papers:
+                ui.warning('Duplicated citekey {}. Keeping the last one.'.format(k))
             try:
                 papers[k] = Paper(k, b)
                 papers[k].added = datetime.datetime.now()
             except ValueError as e:
-                papers[k] = e
+                error = 'Could not load entry for citekey {} ({}).'.format(k, e)
+                if ignore:
+                    ui.warning(error + _IGNORING_MSG)
+                else:
+                    ui.error(error + _ABORT_USE_IGNORE_MSG)
+                    ui.exit()
     return papers
 
 
@@ -75,20 +96,17 @@ def command(conf, args):
 
     rp = repo.Repository(conf)
     # Extract papers from bib
-    papers = many_from_path(bibpath)
+    papers = many_from_path(ui, bibpath, ignore=args.ignore_malformed)
     keys = args.keys or papers.keys()
     for k in keys:
         p = papers[k]
-        if isinstance(p, Exception):
-            ui.error('Could not load entry for citekey {}.'.format(k))
+        rp.push_paper(p, overwrite=args.overwrite)
+        ui.info('{} imported.'.format(color.dye_out(p.citekey, 'citekey')))
+        docfile = bibstruct.extract_docfile(p.bibdata)
+        if docfile is None:
+            ui.warning("No file for {}.".format(p.citekey))
         else:
-            rp.push_paper(p, overwrite=args.overwrite)
-            ui.info('{} imported.'.format(color.dye_out(p.citekey, 'citekey')))
-            docfile = bibstruct.extract_docfile(p.bibdata)
-            if docfile is None:
-                ui.warning("No file for {}.".format(p.citekey))
-            else:
-                rp.push_doc(p.citekey, docfile, copy=copy)
-                #FIXME should move the file if configured to do so.
+            rp.push_doc(p.citekey, docfile, copy=copy)
+            # FIXME should move the file if configured to do so.
 
     rp.close()
diff --git a/pubs/databroker.py b/pubs/databroker.py
index d529c3e..97c1398 100644
--- a/pubs/databroker.py
+++ b/pubs/databroker.py
@@ -45,7 +45,11 @@ class DataBroker(object):
 
     def pull_bibentry(self, citekey):
         bibdata_raw = self.filebroker.pull_bibfile(citekey)
-        return self.endecoder.decode_bibdata(bibdata_raw)
+        try:
+            return self.endecoder.decode_bibdata(bibdata_raw)
+        except self.endecoder.BibDecodingError as e:
+            e.message = "Unable to decode bibtex for paper {}.".format(citekey)
+            raise e
 
     def push_metadata(self, citekey, metadata):
         metadata_raw = self.endecoder.encode_metadata(metadata)
diff --git a/pubs/endecoder.py b/pubs/endecoder.py
index 8fd6941..259b8d9 100644
--- a/pubs/endecoder.py
+++ b/pubs/endecoder.py
@@ -66,6 +66,16 @@ class EnDecoder(object):
         * encode_bibdata will try to recognize exceptions
     """
 
+    class BibDecodingError(Exception):
+
+        message = "Could not parse provided bibdata:\n---\n{}\n---"
+
+        def __init__(self, bibdata):
+            self.data = bibdata
+
+        def __str__(self):
+            return self.message.format(self.data)
+
     bwriter = bp.bwriter.BibTexWriter()
     bwriter.display_order = BIBFIELD_ORDER
 
@@ -103,7 +113,10 @@ class EnDecoder(object):
         return entry
 
     def decode_bibdata(self, bibdata):
-        """"""
+        """Decodes bibdata from string.
+
+        If the decoding fails, returns a BibParseError.
+        """
         try:
             entries = bp.bparser.BibTexParser(
                 bibdata, common_strings=True,
@@ -121,4 +134,5 @@ class EnDecoder(object):
         except Exception:
             import traceback
             traceback.print_exc()
-        raise ValueError('could not parse provided bibdata:\n{}'.format(bibdata))
+        raise self.BibDecodingError(bibdata)
+        # TODO: filter exceptions from pyparsing and pass reason upstream
diff --git a/pubs/uis.py b/pubs/uis.py
index afd90fe..9a39b32 100644
--- a/pubs/uis.py
+++ b/pubs/uis.py
@@ -119,7 +119,7 @@ class InputUI(PrintUI):
         except EOFError:
             self.error('Standard input ended while waiting for answer.')
             self.exit(1)
-        return ustr(data) #.decode('utf-8')
+        return ustr(data)  #.decode('utf-8')
 
     def input_choice_ng(self, options, option_chars=None, default=None, question=''):
         """Ask the user to chose between a set of options. The user is asked
diff --git a/tests/fake_env.py b/tests/fake_env.py
index 1a1fe6c..d099a79 100644
--- a/tests/fake_env.py
+++ b/tests/fake_env.py
@@ -77,6 +77,16 @@ class FakeInput():
             if md.__name__ == 'pubs.uis':
                 md.InputUI.editor_input = self
                 md.InputUI.edit_file = self.input_to_file
+                # Do not catch UnexpectedInput
+                original_handler = md.InputUI.handle_exception
+
+                def handler(ui, exc):
+                    if isinstance(exc, self.UnexpectedInput):
+                        raise
+                    else:
+                        original_handler(ui, exc)
+
+                md.InputUI.handle_exception = handler
 
     def input_to_file(self, path_to_file, temporary=True):
         content.write_file(path_to_file, self())
diff --git a/tests/str_fixtures.py b/tests/str_fixtures.py
index 03360b8..3b6ded3 100644
--- a/tests/str_fixtures.py
+++ b/tests/str_fixtures.py
@@ -69,7 +69,7 @@ bibtex_no_citekey = """@Manual{,
 }
 """
 
-bibtex_month= """@inproceedings{Goyal2017,
+bibtex_month = """@inproceedings{Goyal2017,
  author = {Goyal, Anirudh and Sordoni, Alessandro and C{\^{o}}t{\'{e}}, Marc-Alexandre and Ke, Nan Rosemary and Bengio, Yoshua},
  title = {Z-Forcing: Training Stochastic Recurrent Networks},
  year = {2017},
@@ -78,6 +78,12 @@ bibtex_month= """@inproceedings{Goyal2017,
 }
 """
 
+not_bibtex = """@misc{this looks,
+  like = a = bibtex file but
+    , is not a real one!
+
+"""
+
 sample_conf = """
 [main]
 
diff --git a/tests/test_apis.py b/tests/test_apis.py
index 087f32f..a83481a 100644
--- a/tests/test_apis.py
+++ b/tests/test_apis.py
@@ -32,7 +32,7 @@ class TestDOI2Bibtex(unittest.TestCase):
 
     def test_parse_fails_on_incorrect_DOI(self):
         bib = doi2bibtex('999999')
-        with self.assertRaises(ValueError):
+        with self.assertRaises(EnDecoder.BibDecodingError):
             self.endecoder.decode_bibdata(bib)
 
 
@@ -56,7 +56,7 @@ class TestISBN2Bibtex(unittest.TestCase):
 
     def test_parse_fails_on_incorrect_ISBN(self):
         bib = doi2bibtex('9' * 13)
-        with self.assertRaises(ValueError):
+        with self.assertRaises(EnDecoder.BibDecodingError):
             self.endecoder.decode_bibdata(bib)
 
 
diff --git a/tests/test_endecoder.py b/tests/test_endecoder.py
index 80a9f53..a6e3043 100644
--- a/tests/test_endecoder.py
+++ b/tests/test_endecoder.py
@@ -52,7 +52,7 @@ class TestEnDecode(unittest.TestCase):
 
         self.assertEqual(bibraw1, bibraw2)
 
-    def test_endecode_bibtex(self):
+    def test_endecode_bibtex_converts_month_string(self):
         """Test if `month=dec` is correctly recognized and transformed into
         `month={December}`"""
         decoder = endecoder.EnDecoder()
@@ -147,6 +147,11 @@ class TestEnDecode(unittest.TestCase):
         self.assertIn('author', entry1)
         self.assertIn('institution', entry1)
 
+    def test_endecodes_raises_exception(self):
+        decoder = endecoder.EnDecoder()
+        with self.assertRaises(decoder.BibDecodingError):
+            decoder.decode_bibdata("@misc{I am not a correct bibtex{{}")
+
 
 if __name__ == '__main__':
     unittest.main()
diff --git a/tests/test_usecase.py b/tests/test_usecase.py
index ae98258..43cbf2c 100644
--- a/tests/test_usecase.py
+++ b/tests/test_usecase.py
@@ -128,7 +128,7 @@ class CommandTestCase(fake_env.TestFakeFs):
                         outs.append(color.undye(actual_out))
                     else:
                         pubs_cmd.execute(actual_cmd.split())
-                except fake_env.FakeInput.UnexpectedInput as e:
+                except fake_env.FakeInput.UnexpectedInput:
                     self.fail('Unexpected input asked by command: {}.'.format(
                         actual_cmd))
             return outs
@@ -197,7 +197,7 @@ class TestAlone(CommandTestCase):
     def test_alone_misses_command(self):
         with self.assertRaises(FakeSystemExit) as cm:
             self.execute_cmds(['pubs'])
-            self.assertEqual(cm.exception.code, 2)
+        self.assertEqual(cm.exception.code, 2)
 
     def test_alone_prints_help(self):
         # capturing the output of `pubs --help` is difficult because argparse
@@ -334,11 +334,20 @@ class TestAdd(URLContentTestCase):
     def test_add_no_citekey_fails(self):
         # See #113
         cmds = ['pubs init',
-                ('pubs add', [str_fixtures.bibtex_no_citekey]),
+                ('pubs add', [str_fixtures.bibtex_no_citekey, 'n']),
                 ]
         with self.assertRaises(FakeSystemExit):
             self.execute_cmds(cmds)
 
+    def test_add_edit_fails(self):
+        cmds = ['pubs init',
+                ('pubs add',
+                    ['@misc{I am not a correct bibtex{{}', 'n']),
+                ]
+        with self.assertRaises(FakeSystemExit) as cm:
+            self.execute_cmds(cmds)
+        self.assertEqual(cm.exception.code, 1)
+
 
 class TestList(DataCommandTestCase):
 
@@ -679,16 +688,40 @@ class TestUsecase(DataCommandTestCase):
         self.assertFileContentEqual(os.path.expanduser('~/.pubsrc'),
                                     str_fixtures.sample_conf)
 
-    def test_editor_abort(self):
+    def test_editor_aborts(self):
         with self.assertRaises(FakeSystemExit):
             cmds = ['pubs init',
-                    ('pubs add', ['abc', 'n']),
-                    ('pubs add', ['abc', 'y', 'abc', 'n']),
                     'pubs add data/pagerank.bib',
-                    ('pubs edit Page99', ['', 'a']),
+                    ('pubs edit Page99', ['', 'n']),
                    ]
             self.execute_cmds(cmds)
 
+    def test_editor_succeeds_on_second_edit(self):
+        cmds = ['pubs init',
+                'pubs add data/pagerank.bib',
+                ('pubs edit Page99', [
+                    '@misc{Page99, title="TTT" author="X. YY"}', 'y',
+                    '@misc{Page99, title="TTT", author="X. YY"}', '']),
+                ('pubs list', [], '[Page99] YY, X. "TTT" \n')
+                ]
+        self.execute_cmds(cmds)
+
+    def test_add_aborts(self):
+        with self.assertRaises(FakeSystemExit):
+            cmds = ['pubs init',
+                    ('pubs add New', ['']),
+                   ]
+            self.execute_cmds(cmds)
+
+    def test_add_succeeds_on_second_edit(self):
+        cmds = ['pubs init',
+                ('pubs add', [
+                    '', 'y',
+                    '@misc{New, title="TTT", author="X. YY"}', '']),
+                ('pubs list', [], '[New] YY, X. "TTT" \n')
+                ]
+        self.execute_cmds(cmds)
+
     def test_editor_success(self):
         cmds = ['pubs init',
                 ('pubs add', [str_fixtures.bibtex_external0]),
@@ -790,6 +823,25 @@ class TestUsecase(DataCommandTestCase):
         outs = self.execute_cmds(cmds)
         self.assertEqual(1 + 1, len(outs[-1].split('\n')))
 
+    def test_import_fails_without_ignore(self):
+        with FakeFileOpen(self.fs)('data/fake.bib', 'w') as f:
+            f.write(str_fixtures.not_bibtex)
+        cmds = ['pubs init',
+                'pubs import data/ Page99',
+                ]
+        with self.assertRaises(FakeSystemExit):
+            self.execute_cmds(cmds)
+
+    def test_import_ignores(self):
+        with FakeFileOpen(self.fs)('data/fake.bib', 'w') as f:
+            f.write(str_fixtures.not_bibtex)
+        cmds = ['pubs init',
+                'pubs import --ignore-malformed data/ Page99',
+                'pubs list'
+                ]
+        outs = self.execute_cmds(cmds)
+        self.assertEqual(1 + 1, len(outs[-1].split('\n')))
+
     def test_update(self):
         cmds = ['pubs init',
                 'pubs add data/pagerank.bib',