USPTO XML format causes BaseX GUI errors

On Sun, Apr 27, 2014 at 6:52 PM, Jose Rey <onlyrey at gmail.com> wrote: > Hi, > > I am new to BaseX and will attempt to use it to analyze xml datasets from > R at some point in the near future. I am using a BaseX GUI under Windows-7 > operating system and had an error while trying to create a database using > the GUI by using an XML file as input. The file comes from the US Patents > and Trademarks Office (USPTO), and the larger XML datasets they provide > have the same problem. > > The error text is: > > Command: > CREATE DB ipgb20110104Sample C:/Users/admin/Downloads/ipgb20110104Sample.xmlError:"C:/Users/admin/Downloads/ipgb20110104Sample.xml" (Line 306): The processing instruction target matching "[xX][mM][lL]" is not allowed. > > > The XML document I am trying to open is contained in this zip file: > http://www.uspto.gov/products/ipgb110104-sample.zip > > The link to the document is in this page: > http://www.uspto.gov/products/xml-resources.jsp > under the "Patent Grant Data / XML ST. 36 (ICE) v4.2 (a.k.a. Red Book) > (2007 - 2012)" section of the page, under the "Sample Documents > (Bibliographic)" bulletpoint > > > From searching online, I found that the error is because of poor > formatting in the file. However, larger datasets of the same kind (USPTO > bulk download @ Google) have the same problem. In the specific case of the > file I mention, it has a carriage return in the first line, and then has > several concatenated XML files, which is the case of the larger XML files > from the USPTO. > > My question is: > > Is there a work around to this error/problem? Can I somehow tell BaseX to > ignore or somehow acknowledge that mistake and load the file(s). > > Thank you so much, > > Jose > > -- > Jose I. Rey > onlyrey at gmail.com >

On Sun, Apr 27, 2014 at 1:19 PM, Christian Grün <christian.gruen at gmail.com> wrote: > Hi Jose, > > your .xml file actually contains several XML files, which must first be > split in order to be parsed. I’ve attached one solution in XQuery (there > may be other, more elegant solutions): > > (: open file :) > let $input := unparsed-text('ipgb20110104Sample.xml') > (: get document substrings; omit those without angle brackets :) > let $docs := tokenize($input, '<\?xml version="1.0" > encoding="UTF-8"\?>\s*<!DOCTYPE us-patent-grant SYSTEM > "us-patent-grant-v42-2006-08-23.dtd" \[ \]>\s*')[matches(., '<')] > (: generate document names :) > let $names := > for $n in 1 to count($docs) > return $n || '.xml' > (: create database with all documents :) > return db:create('db', $docs, $names) > > Hope this helps, > Christian > > > > > On Sun, Apr 27, 2014 at 6:52 PM, Jose Rey <onlyrey at gmail.com> wrote: > >> Hi, >> >> I am new to BaseX and will attempt to use it to analyze xml datasets from >> R at some point in the near future. I am using a BaseX GUI under Windows-7 >> operating system and had an error while trying to create a database using >> the GUI by using an XML file as input. The file comes from the US Patents >> and Trademarks Office (USPTO), and the larger XML datasets they provide >> have the same problem. >> >> The error text is: >> >> Command: >> CREATE DB ipgb20110104Sample C:/Users/admin/Downloads/ipgb20110104Sample.xmlError:"C:/Users/admin/Downloads/ipgb20110104Sample.xml" (Line 306): The processing instruction target matching "[xX][mM][lL]" is not allowed. >> >> >> The XML document I am trying to open is contained in this zip file: >> http://www.uspto.gov/products/ipgb110104-sample.zip >> >> The link to the document is in this page: >> http://www.uspto.gov/products/xml-resources.jsp >> under the "Patent Grant Data / XML ST. 36 (ICE) v4.2 (a.k.a. Red Book) >> (2007 - 2012)" section of the page, under the "Sample Documents >> (Bibliographic)" bulletpoint >> >> >> From searching online, I found that the error is because of poor >> formatting in the file. However, larger datasets of the same kind (USPTO >> bulk download @ Google) have the same problem. In the specific case of the >> file I mention, it has a carriage return in the first line, and then has >> several concatenated XML files, which is the case of the larger XML files >> from the USPTO. >> >> My question is: >> >> Is there a work around to this error/problem? Can I somehow tell BaseX to >> ignore or somehow acknowledge that mistake and load the file(s). >> >> Thank you so much, >> >> Jose >> >> -- >> Jose I. Rey >> onlyrey at gmail.com >> > >