Fix #66

Concatenate 'title' and 'id' of a product to build it's name.
Add a 'sanitize' method to remove spaces, punctuation,.... in product name

eodag/plugins/download/http.py

@@ -29,7 +29,7 @@
 
 from eodag.plugins.download.base import Download
 from eodag.plugins.search.base import MisconfiguredError
-
+from eodag.utils import sanitize
 
 logger = logging.getLogger('eodag.plugins.download.http')
 
@@ -68,7 +68,12 @@ def download(self, product, auth=None, progress_callback=None):
         # Strong asumption made here: all products downloaded will be zip files
         # If they are not, the '.zip' extension will be removed when they are downloaded and returned as is
         prefix = os.path.abspath(self.config['outputs_prefix'])
-        fs_path = os.path.join(prefix, '{}.zip'.format(product.properties['title']))
+        sanitized_title = sanitize(product.properties['title'])
+        collision_avoidance_suffix = '' \
+            if sanitized_title == product.properties['title'] \
+            else ('-' + sanitize(product.properties['id']))
+        fs_path = os.path.join(prefix, '{}{}.zip'.format(sanitize(product.properties['title']),
+                                                         collision_avoidance_suffix))
         download_records_dir = os.path.join(prefix, '.downloaded')
         try:
             os.makedirs(download_records_dir)

eodag/utils/__init__.py

@@ -18,6 +18,7 @@
 from __future__ import unicode_literals
 
 import re
+import string
 import sys
 import types
 import unicodedata
@@ -28,6 +29,7 @@
 from rasterio.crs import CRS
 from requests.auth import AuthBase
 from tqdm import tqdm, tqdm_notebook
+from unidecode import unidecode
 
 
 # All modules using these should import them from utils package
@@ -119,6 +121,29 @@ def utf8_everywhere(mapping):
     )
 
 
+def sanitize(value):
+    """Sanitize string to be used as a name of a directory.
+    >>> sanitize('productName')
+    'productName'
+    >>> sanitize('name with multiple  spaces')
+    'name_with_multiple_spaces'
+    >>> sanitize('âtre fête île alcôve bûche çà génèse où Noël ovoïde capharnaüm')
+    'atre_fete_ile_alcove_buche_ca_genese_ou_Noel_ovoide_capharnaum'
+    >>> sanitize('replace,ponctuation:;signs!?byunderscorekeeping-hyphen.dot_and_underscore')
+    'replace_ponctuation_signs_byunderscorekeeping-hyphen.dot_and_underscore'
+    """
+    # remove accents
+    rv = unidecode(value)
+    # replace punctuation signs and spaces by underscore
+    ## keep hyphen, dot and underscore from punctuation
+    tobereplaced = re.sub('[-_.]', '', string.punctuation)
+    ## add spaces to be removed
+    tobereplaced += '\s'
+
+    rv = re.sub('[' + tobereplaced + ']+', '_', rv)
+    return str(rv)
+
+
 def mutate_dict_in_place(func, mapping):
     """Apply func to values of mapping.
 

setup.py

@@ -61,6 +61,7 @@
         'jsonpath-rw',
         'lxml',
         'xarray',
+        'unidecode == 1.0.22',
     ],
     extras_require={
         'dev': [