Commit bec6acf4 authored by delanoe's avatar delanoe

[TEXT] Adapting messages for corpus.

parent 82084d13
...@@ -199,7 +199,7 @@ RESOURCETYPES = [ ...@@ -199,7 +199,7 @@ RESOURCETYPES = [
'crawler': None, 'crawler': None,
}, },
{ 'type': 3, { 'type': 3,
'name': 'Pubmed [XML]', 'name': 'Pubmed [CRAWLER/XML]',
'format': 'Pubmed', 'format': 'Pubmed',
'parser': "PubmedParser", 'parser': "PubmedParser",
'file_formats':["zip", "xml"], 'file_formats':["zip", "xml"],
...@@ -235,14 +235,14 @@ RESOURCETYPES = [ ...@@ -235,14 +235,14 @@ RESOURCETYPES = [
'crawler': None, 'crawler': None,
}, },
{ 'type': 8, { 'type': 8,
'name': 'ISTex', 'name': 'ISTex [CRAWLER]',
'format': 'json', 'format': 'json',
'parser': "ISTexParser", 'parser': "ISTexParser",
'file_formats':["zip", "txt"], 'file_formats':["zip", "txt"],
'crawler': None, 'crawler': None,
}, },
{ "type": 9, { "type": 9,
"name": 'SCOAP [XML]', "name": 'SCOAP [CRAWLER/XML]',
"parser": "CernParser", "parser": "CernParser",
"format": 'MARC21', "format": 'MARC21',
'file_formats':["zip","xml"], 'file_formats':["zip","xml"],
...@@ -257,7 +257,7 @@ RESOURCETYPES = [ ...@@ -257,7 +257,7 @@ RESOURCETYPES = [
# }, # },
# #
{ "type": 10, { "type": 10,
"name": 'REPEC [MULTIVAC]', "name": 'REPEC [CRAWLER]',
"parser": "MultivacParser", "parser": "MultivacParser",
"format": 'JSON', "format": 'JSON',
'file_formats':["zip","json"], 'file_formats':["zip","json"],
......
...@@ -89,13 +89,15 @@ class MultivacCrawler(Crawler): ...@@ -89,13 +89,15 @@ class MultivacCrawler(Crawler):
corpus = [] corpus = []
paging = 100 paging = 100
self.query_max = self.scan_results(query) self.query_max = self.scan_results(query)
print("self.query_max : %s" % self.query_max)
if self.query_max > QUERY_SIZE_N_MAX: if self.query_max > QUERY_SIZE_N_MAX:
msg = "Invalid sample size N = %i (max = %i)" % (self.query_max, QUERY_SIZE_N_MAX) msg = "Invalid sample size N = %i (max = %i)" % (self.query_max, QUERY_SIZE_N_MAX)
print("ERROR (scrap: multivac d/l ): ",msg) print("ERROR (scrap: Multivac d/l ): ",msg)
self.query_max = QUERY_SIZE_N_MAX self.query_max = QUERY_SIZE_N_MAX
for page in range(1, trunc(self.query_max / 100) + 1): for page in range(1, trunc(self.query_max / 100) + 1):
print("Downloading page %s to %s results" % (page, paging))
docs = self._get(query, fromPage=page, count=paging)["results"]["hits"] docs = self._get(query, fromPage=page, count=paging)["results"]["hits"]
for doc in docs: for doc in docs:
corpus.append(doc) corpus.append(doc)
......
...@@ -86,12 +86,12 @@ ...@@ -86,12 +86,12 @@
<button type="button" class="close" data-dismiss="modal" aria-label="Close"> <button type="button" class="close" data-dismiss="modal" aria-label="Close">
<span aria-hidden="true">&times;</span> <span aria-hidden="true">&times;</span>
</button> </button>
<h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span> Uploading corpus...</h2> <h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span>Building corpus...</h2>
</div> </div>
<div class="modal-body"> <div class="modal-body">
<h5> <h5>
Your file has been uploaded ! Gargantext is gathering your texts
Gargantext need some time to eat it. and need some time to eat it.
Duration depends on the size of the dish. Duration depends on the size of the dish.
</h5> </h5>
</div> </div>
......
...@@ -322,12 +322,12 @@ ...@@ -322,12 +322,12 @@
<button type="button" class="close" data-dismiss="modal" aria-label="Close"> <button type="button" class="close" data-dismiss="modal" aria-label="Close">
<span aria-hidden="true">&times;</span> <span aria-hidden="true">&times;</span>
</button> </button>
<h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span> Uploading corpus...</h2> <h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span> Building the corpus...</h2>
</div> </div>
<div class="modal-body"> <div class="modal-body">
<p> <p>
Your file has been uploaded ! Gargantext is gathering your texts
Gargantext need some time to eat it. and need some time to eat it.
Duration depends on the size of the dish. Duration depends on the size of the dish.
</p> </p>
</div> </div>
......
...@@ -199,12 +199,12 @@ ...@@ -199,12 +199,12 @@
<button type="button" class="close" data-dismiss="modal" aria-label="Close"> <button type="button" class="close" data-dismiss="modal" aria-label="Close">
<span aria-hidden="true">&times;</span> <span aria-hidden="true">&times;</span>
</button> </button>
<h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span> Uploading corpus...</h2> <h2 class="modal-title"><h2><span class="glyphicon glyphicon-info-sign" aria-hidden="true"></span>Building the corpus...</h2>
</div> </div>
<div class="modal-body"> <div class="modal-body">
<p> <p>
Your file has been uploaded ! Gargantext is gathering your texts
Gargantext need some time to eat it. and need some time to eat it.
Duration depends on the size of the dish. Duration depends on the size of the dish.
</p> </p>
</div> </div>
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment