Attachment Collection

The attachment collection is a separate index in CLAIMS Direct which is accessed through the standard /search/query API endpoint using the parameter qt=attachments. However, the Solr documents returned are in a different format from the normal index. In the normal index, the Solr document represents a patent publication. In the attachment index, the Solr document represents a specific attachment file. The ramifications of this are that

q=ucid:EP-0700000-B1

returns 6 results instead of 1 result with 6 attachment fields.

You can also combine path and fname to create the value that will retrieve an attachment from /attachment/fetch. For example, given the values:

               fnam : "US20170077426A1-20170316-C00118.MOL"

path : "375962165/US/20170316/A1/020170/07/74/26"

the following request returns the attachment:

/attachment/fetch?path=375962165/US/20170316/A1/020170/07/74/26/US20170077426A1-20170316-C00118.MOL

Because it is a separate index, the attachment collection is not accessible through CDWI.

FieldDescriptionField TypeExamplesMultivaluedIndexedStoredRequired
fnamattachment file namealexandria_tokenUS20170077426A1-20170316-C00118.MOL
xxx
sizesize of the attachment in bytestintFind all attachments larger than 1MB
size:[ 1048576 TO * ]

xxx
(default=1)
pathpath to the attachmentalexandria_token375962165/US/20170316/A1/020170/07/74/26
xxx
sffx

file suffix

Note: See Attachment Server Description and Coverage for values.

string**pdf
x
x
ucidunique patent document identifier based on country, doc-number, and kindstring**

AP-584-A

AP-584*


xxx
pnpublication identifier based on country, publication number, and kindalexandria_token

AP-584-A

AP584A

AP584

AP

B2


x
x
xxpnctrypublication countrystring**US
x
x
xxpnkindpublication kind codestring**

B1

A*


x
x
pd

publication date

Note: If no ./date is present and a document-id element exists, a default value of 00010101 is inserted.

tint20010101
x
x
xxpd_d

publication date represented internally as a date type

Note: This field allows efficient range faceting and Solr date math. Quotes or query escaping may be required for proper searching. See Working with Dates for more information.

tdate

Format: "YYYY-MM-DDThh:mm:ssZ"

"2007-01-01T00:01:01Z"

2007-01-01T00\:01\:01Z

[2017-07-05T00:00:00.000Z TO *]

[NOW-14DAYS TO NOW] (Find documents with publication dates in the last 14 days)


x
x
xxpdyearpublication yeartint2001
x
x
textcombination of ucid, pkey, fnam, sffx, and pathalexandria_tokentext: us AND pdfxx

aidunique attachment identifierstring**N/A – used internally for index updates
x x x
pkeyprimary key of the attachment made up of "ucid-publication date", e.g., US-20170077426-A1-20170316alexandria_tokenN/A – used internally for index updates
x
x
slvlsubscription level identifier of request (0=Basic, 1=Premium, 2=Premium+)tintN/A – used internally for index filtering
xxx

**Note: String fields are not tokenized. Therefore, queries in these fields will only return exact matches unless wild cards are used.

Example

Request

wget -O- -q  \
  --header 'x-user:x' 
  --header 'x-password:y' \
'https://cdws21.ificlaims.com/search/query?qt=attachments&rows=4&q=pnctry:US AND pd:20190312 AND sffx:mol'

Response

{
   "time": "0.021077",
   "status" : "success",
   "content": {
     "responseHeader" : {
       "params" : {
         "timeAllowed" : "300000",
         "wt" : "json",
         "q" : "pnctry:US AND pd:20190312 AND sffx:mol",
         "fq" : "",
         "rows" : "0",
         "indent" : "true",
         "start" : "0"
       },
       "status" : 0,
       "QTime" : 1
     },
     "response" : {
       "numFound" : 24179,
       "docs" : [],
       "start" : 0
      }
    }
  }