Loading ...
Sorry, an error occurred while loading the content.

Re: [govtrack] 2 issues

Expand Messages
  • Ben Woosley
    Thanks Josh, On rolls, you might consider using the new, year-less scheme across the board, with soft links or the like aliasing the xml to the old naming
    Message 1 of 14 , May 2, 2010
    • 0 Attachment
      Thanks Josh,

      On rolls, you might consider using the new, year-less scheme across the board, with soft links or the like aliasing the xml to the old naming scheme.

      -Ben


      On Sun, May 2, 2010 at 11:32 AM, Josh Tauberer <tauberer@...> wrote:
         * Rolls from the keithpoole data source use a different naming

           scheme than we see elsewhere.  e.g. 'h16.xml' rather than
           '1987-h16.xml'. For congress 101, both sources are present, under
           the 2 different naming schemes.

      Right I was forced into this because earlier in Congressional history (esp at the beginning) Congresses didn't line up with years very nicely. In fact, they still don't but as a matter of practicality it hadn't mattered. In the 1st Congress, for instance, it ends on March 3, 1791 and the 2nd Congress starts the next day. That means a vote with an id that has the year and roll call number is ambiguous between possibly two sessions.

      So I changed the naming scheme for roll call votes brought in this way to just the roll call number in the files, and in the website URLs I think it has the session number and roll call number.

      An unavoidable mess.

      In the 101st Congress, I think the Senate votes may go with the new (i.e. 102-111th) scheme and the House votes may go with the historical (i.e. 1st-101st) scheme. Maybe.

         * Most senate bills missing for congress 101.  This is a bit odd

           because the roll data seems complete, while on the Blll side,
           there are many missing bills, including it seems the vast majority
           of Senate bills.  Congress 100 and 102 don't seem to have this problem

      I'm not sure how far back I've imported data from THOMAS. The further I go back, the more errors pop up. For sure it's relatively complete as far back as the 103rd. Also, bill and vote data come from different sources, so you can't judge the completeness of one by the other.

      These things have all been in progress for a long time...

      Bobby Jindal carries a PVS id (26879) which has apparently been retired and replaced (with 35481).

      I'll update the db. I used to enter PVS IDs myself but now get it from Sunlight's API on a periodical/random basis.

      Thanks.

      Josh

    • Josh Tauberer
      ... Ah but the House and Senate currently reset their numbering at the start of each calendar year, so e.g. there are two roll #1 s in each chamber for the
      Message 2 of 14 , May 11, 2010
      • 0 Attachment
        On 05/02/2010 08:14 PM, Ben Woosley wrote:
        > On rolls, you might consider using the new, year-less scheme across the
        > board, with soft links or the like aliasing the xml to the old naming
        > scheme.

        Ah but the House and Senate currently reset their numbering at the start
        of each calendar year, so e.g. there are two roll #1's in each chamber
        for the 111th Congress.

        (What's a poor schema designer to do? :)

        - Josh Tauberer
        - CivicImpulse / GovTrack.us

        http://razor.occams.info | www.govtrack.us | civicimpulse.com

        "Members of both sides are reminded not to use guests of the
        House as props."

        On 05/02/2010 08:14 PM, Ben Woosley wrote:
        >
        >
        > Thanks Josh,
        >
        > On rolls, you might consider using the new, year-less scheme across the
        > board, with soft links or the like aliasing the xml to the old naming
        > scheme.
        >
        > -Ben
        >
        >
        > On Sun, May 2, 2010 at 11:32 AM, Josh Tauberer <tauberer@...
        > <mailto:tauberer@...>> wrote:
        >
        > * Rolls from the keithpoole data source use a different naming
        >
        > scheme than we see elsewhere. e.g. 'h16.xml' rather than
        > '1987-h16.xml'. For congress 101, both sources are present, under
        > the 2 different naming schemes.
        >
        >
        > Right I was forced into this because earlier in Congressional
        > history (esp at the beginning) Congresses didn't line up with years
        > very nicely. In fact, they still don't but as a matter of
        > practicality it hadn't mattered. In the 1st Congress, for instance,
        > it ends on March 3, 1791 and the 2nd Congress starts the next day.
        > That means a vote with an id that has the year and roll call number
        > is ambiguous between possibly two sessions.
        >
        > So I changed the naming scheme for roll call votes brought in this
        > way to just the roll call number in the files, and in the website
        > URLs I think it has the session number and roll call number.
        >
        > An unavoidable mess.
        >
        > In the 101st Congress, I think the Senate votes may go with the new
        > (i.e. 102-111th) scheme and the House votes may go with the
        > historical (i.e. 1st-101st) scheme. Maybe.
        >
        > * Most senate bills missing for congress 101. This is a bit odd
        >
        > because the roll data seems complete, while on the Blll side,
        > there are many missing bills, including it seems the vast
        > majority
        > of Senate bills. Congress 100 and 102 don't seem to have
        > this problem
        >
        >
        > I'm not sure how far back I've imported data from THOMAS. The
        > further I go back, the more errors pop up. For sure it's relatively
        > complete as far back as the 103rd. Also, bill and vote data come
        > from different sources, so you can't judge the completeness of one
        > by the other.
        >
        > These things have all been in progress for a long time...
        >
        > Bobby Jindal carries a PVS id (26879) which has apparently been
        > retired and replaced (with 35481).
        >
        >
        > I'll update the db. I used to enter PVS IDs myself but now get it
        > from Sunlight's API on a periodical/random basis.
        >
        > Thanks.
        >
        > Josh
        >
        >
        >
        >
        >
      • Harvey Frey
        Has anyone tried downloading bills from Thomas in HTML format, or converting their Text downloads to HTML? The guy at the Thomas help desk at first didn t
        Message 3 of 14 , May 24, 2010
        • 0 Attachment
              Has anyone tried downloading bills from Thomas in HTML format, or converting their Text downloads to HTML?

              The guy at the Thomas help desk at first didn't understand what I wanted, and then said that no one had ever asked for that before.


              As you know, when you download the Contents page of a bill, the hyperlinks to the actual sections point to your own local folder instead of to the Thomas page where they exist.

              If you add a base statement to point them back to the Thomas site, the links disappear in a few minutes since the search expires.

              If you download all their referenced pages, you can't use their links to convert them to local hyperlinks, since they address them through a cgi program instead of through static name anchors.

              Your can download a bill as plain text rather than HTML, but manually adding all the name anchors and hyperlinks would be a massive job.

              (I did it for the Patriot Act, and it wasn't fun, especially since it looked like they intentionally obfuscated the references, sometimes using Public Law References, sometimes USC, and sometimes common names, so it was a detective job to find the sections they were amending.)

              Has anyone tried this, say with perl?

              I'm specifically interested in HR 3590, the recent Health Reform Bill.
          Some bills are posted as a massive XML file, but this one isn't. (If it were, I suppose you could use their id/id-ref pairs to construct href/name pairs, and then clean out the rest of the XML cruft.)

          Harvey

          =============================
          Harvey S. Frey MD PhD Esq.
          hsfrey@...  www.harp.org
          -----------------------------
          "Withdrawing in disgust is not the same thing as apathy."
          - Brian Eno

          =============================
        • Josh Tauberer
          That s what GovTrack does to get bill text. http://www.govtrack.us/data/us/bills.text/111/h/h3590.html I clean the HTML and make sure it s well-formed XML
          Message 4 of 14 , May 24, 2010
          • 0 Attachment
            That's what GovTrack does to get bill text.
            http://www.govtrack.us/data/us/bills.text/111/h/h3590.html

            I clean the HTML and make sure it's well-formed XML before putting it there.

            - Josh Tauberer
            - CivicImpulse / GovTrack.us

            http://razor.occams.info | www.govtrack.us | civicimpulse.com

            "Members of both sides are reminded not to use guests of the
            House as props."

            On 05/24/2010 06:59 PM, Harvey Frey wrote:
            >
            >
            > Has anyone tried downloading bills from Thomas in HTML format, or
            > converting their Text downloads to HTML?
            >
            > The guy at the Thomas help desk at first didn't understand what I
            > wanted, and then said that no one had ever asked for that before.
            >
            > As you know, when you download the Contents page of a bill, the
            > hyperlinks to the actual sections point to your own local folder instead
            > of to the Thomas page where they exist.
            >
            > If you add a base statement to point them back to the Thomas site, the
            > links disappear in a few minutes since the search expires.
            >
            > If you download all their referenced pages, you can't use their links to
            > convert them to local hyperlinks, since they address them through a cgi
            > program instead of through static name anchors.
            >
            > Your can download a bill as plain text rather than HTML, but manually
            > adding all the name anchors and hyperlinks would be a massive job.
            >
            > (I did it for the Patriot Act, and it wasn't fun, especially since it
            > looked like they intentionally obfuscated the references, sometimes
            > using Public Law References, sometimes USC, and sometimes common names,
            > so it was a detective job to find the sections they were amending.)
            >
            > Has anyone tried this, say with perl?
            >
            > I'm specifically interested in HR 3590, the recent Health Reform Bill.
            > Some bills are posted as a massive XML file, but this one isn't. (If it
            > were, I suppose you could use their id/id-ref pairs to construct
            > href/name pairs, and then clean out the rest of the XML cruft.)
            >
            > Harvey
            >
            > =============================
            > Harvey S. Frey MD PhD Esq.
            > hsfrey@... www.harp.org
            > -----------------------------
            > "Withdrawing in disgust is not the same thing as apathy."
            > - Brian Eno
            > =============================
            >
            >
            >
          • Neil M.
            I ve got an old python script that takes the US Code text files (from http://uscode.house.gov), scans for references and outputs mediawiki formatted pages and
            Message 5 of 14 , May 24, 2010
            • 0 Attachment
              I've got an old python script that takes the US Code text files (from
              http://uscode.house.gov), scans for references and outputs mediawiki
              formatted pages and wiki links. It shouldn't be too difficult to modify
              it to parse some other text/HTML file and output HTML instead. What is
              it you want to link to exactly? The aforementioned US Code website?
              Thomas? Both?

              Neil

              On 5/24/2010 3:59 PM, Harvey Frey wrote:
              >
              >
              > Has anyone tried downloading bills from Thomas in HTML format, or
              > converting their Text downloads to HTML?
              >
              > The guy at the Thomas help desk at first didn't understand what I
              > wanted, and then said that no one had ever asked for that before.
              >
              > As you know, when you download the Contents page of a bill, the
              > hyperlinks to the actual sections point to your own local folder instead
              > of to the Thomas page where they exist.
              >
              > If you add a base statement to point them back to the Thomas site,
              > the links disappear in a few minutes since the search expires.
              >
              > If you download all their referenced pages, you can't use their
              > links to convert them to local hyperlinks, since they address them
              > through a cgi program instead of through static name anchors.
              >
              > Your can download a bill as plain text rather than HTML, but
              > manually adding all the name anchors and hyperlinks would be a massive job.
              >
              > (I did it for the Patriot Act, and it wasn't fun, especially since
              > it looked like they intentionally obfuscated the references, sometimes
              > using Public Law References, sometimes USC, and sometimes common names,
              > so it was a detective job to find the sections they were amending.)
              >
              > Has anyone tried this, say with perl?
              >
              > I'm specifically interested in HR 3590, the recent Health Reform
              > Bill. Some bills are posted as a massive XML file, but this one isn't.
              > (If it were, I suppose you could use their id/id-ref pairs to construct
              > href/name pairs, and then clean out the rest of the XML cruft.)
              >
              > Harvey
              >
              > =============================
              > Harvey S. Frey MD PhD Esq.
              > hsfrey@... www.harp.org
              > -----------------------------
              > "Withdrawing in disgust is not the same thing as apathy."
              > - Brian Eno
              > =============================
              >
              >
            • Harvey Frey
              Josh: What I m trying to do is clean it up so the section hierarchy is displayed, and add hyperlinks to the text from the TOC. This is an excerpt of the TOC.
              Message 6 of 14 , May 24, 2010
              • 0 Attachment
                Josh:

                    What I'm trying to do is clean it up so the section hierarchy is displayed, and add hyperlinks to the text from the TOC. This is an excerpt of the TOC. What doesn't show here is that the lines are hyperlinks, and I've used a different background color for the sections of text added to other acts. (The back-ticks are not very salient.)

                    It's not that I'm a neat freak - I just get easily confused when the margins are all raggedy and the fonts don't go with the hierarchy.

                Harvey

                SECTION 1. SHORT TITLE; TABLE OF CONTENTS.

                  (a) Short Title- This Act may be cited as the `Patient Protection and Affordable Care Act'.
                  (b) Table of Contents- The table of contents of this Act is as follows:
                    Sec. 1. Short title; table of contents.

                TITLE I--QUALITY, AFFORDABLE HEALTH CARE FOR ALL AMERICANS



                Josh Tauberer wrote:
                 

                That's what GovTrack does to get bill text.
                http://www.govtrack.us/data/us/bills.text/111/h/h3590.html

                I clean the HTML and make sure it's well-formed XML before putting it there.

                - Josh Tauberer
                - CivicImpulse / GovTrack.us

                http://razor.occams.info | www.govtrack.us | civicimpulse.com

                "Members of both sides are reminded not to use guests of the
                House as props."

                On 05/24/2010 06:59 PM, Harvey Frey wrote:
                >
                >
                > Has anyone tried downloading bills from Thomas in HTML format, or
                > converting their Text downloads to HTML?
                >
                > The guy at the Thomas help desk at first didn't understand what I
                > wanted, and then said that no one had ever asked for that before.
                >
                > As you know, when you download the Contents page of a bill, the
                > hyperlinks to the actual sections point to your own local folder instead
                > of to the Thomas page where they exist.
                >
                > If you add a base statement to point them back to the Thomas site, the
                > links disappear in a few minutes since the search expires.
                >
                > If you download all their referenced pages, you can't use their links to
                > convert them to local hyperlinks, since they address them through a cgi
                > program instead of through static name anchors.
                >
                > Your can download a bill as plain text rather than HTML, but manually
                > adding all the name anchors and hyperlinks would be a massive job.
                >
                > (I did it for the Patriot Act, and it wasn't fun, especially since it
                > looked like they intentionally obfuscated the references, sometimes
                > using Public Law References, sometimes USC, and sometimes common names,
                > so it was a detective job to find the sections they were amending.)
                >
                > Has anyone tried this, say with perl?
                >
                > I'm specifically interested in HR 3590, the recent Health Reform Bill.
                > Some bills are posted as a massive XML file, but this one isn't. (If it
                > were, I suppose you could use their id/id-ref pairs to construct
                > href/name pairs, and then clean out the rest of the XML cruft.)
                >
                > Harvey
                >
                > =============================
                > Harvey S. Frey MD PhD Esq.
                > hsfrey@... www.harp.org
                > -----------------------------
                > "Withdrawing in disgust is not the same thing as apathy."
                > - Brian Eno
                > =============================
                >
                >
                >

              • Harvey Frey
                Hi Neil: Thanks for the response! I simply want to use the contents section of a bill to hyperlink to the actual text paragraphs within the same bill, to
                Message 7 of 14 , May 25, 2010
                • 0 Attachment
                  Hi Neil:

                      Thanks for the response!

                      I simply want to use the 'contents' section of a bill to hyperlink to the actual text paragraphs within the same bill, to make the bills easier to navigate and comprehend.

                      If I save the contents page from Thomas, the links point to a cgi script and disappear within a few minutes. AFAIK, Thomas uses no permanent links - everything runs through their cgi script.

                      If I download the entire bill from Thomas, it contains no links at all, so I need to be able to find section headings and put name anchors there, and put corresponding href anchors in the correct TOC line, but not do it for incidental references which are not headings.

                      So it's not a problem of finding stereotypical text references to USC sections and constructing HTML from them. That would be a pretty straightforward RegExp problem.

                  Harvey
                  ================================================

                  Neil M. wrote:
                  I've got an old python script that takes the US Code text files (from
                  http://uscode.house.gov), scans for references and outputs mediawiki
                  formatted pages and wiki links.  It shouldn't be too difficult to modify
                  it to parse some other text/HTML file and output HTML instead.  What is
                  it you want to link to exactly?  The aforementioned US Code website?
                  Thomas?  Both?
                  
                  Neil
                  
                  On 5/24/2010 3:59 PM, Harvey Frey wrote:
                    
                   
                  
                      Has anyone tried downloading bills from Thomas in HTML format, or
                  converting their Text downloads to HTML?
                  
                      The guy at the Thomas help desk at first didn't understand what I
                  wanted, and then said that no one had ever asked for that before.
                  
                      As you know, when you download the Contents page of a bill, the
                  hyperlinks to the actual sections point to your own local folder instead
                  of to the Thomas page where they exist.
                  
                      If you add a base statement to point them back to the Thomas site,
                  the links disappear in a few minutes since the search expires.
                  
                      If you download all their referenced pages, you can't use their
                  links to convert them to local hyperlinks, since they address them
                  through a cgi program instead of through static name anchors.
                  
                      Your can download a bill as plain text rather than HTML, but
                  manually adding all the name anchors and hyperlinks would be a massive job.
                  
                      (I did it for the Patriot Act, and it wasn't fun, especially since
                  it looked like they intentionally obfuscated the references, sometimes
                  using Public Law References, sometimes USC, and sometimes common names,
                  so it was a detective job to find the sections they were amending.)
                  
                      Has anyone tried this, say with perl?
                  
                      I'm specifically interested in HR 3590, the recent Health Reform
                  Bill. Some bills are posted as a massive XML file, but this one isn't.
                  (If it were, I suppose you could use their id/id-ref pairs to construct
                  href/name pairs, and then clean out the rest of the XML cruft.)
                  
                  Harvey
                  
                  =============================
                  Harvey S. Frey MD PhD Esq.
                  hsfrey@...  www.harp.org
                  -----------------------------
                  "Withdrawing in disgust is not the same thing as apathy."
                  - Brian Eno
                  =============================
                  
                  
                      
                  
                  ------------------------------------
                  
                  Yahoo! Groups Links
                  
                  <*> To visit your group on the web, go to:
                      http://groups.yahoo.com/group/govtrack/
                  
                  <*> Your email settings:
                      Individual Email | Traditional
                  
                  <*> To change settings online go to:
                      http://groups.yahoo.com/group/govtrack/join
                      (Yahoo! ID required)
                  
                  <*> To change settings via email:
                      govtrack-digest@yahoogroups.com 
                      govtrack-fullfeatured@yahoogroups.com
                  
                  <*> To unsubscribe from this group, send an email to:
                      govtrack-unsubscribe@yahoogroups.com
                  
                  <*> Your use of Yahoo! Groups is subject to:
                      http://docs.yahoo.com/info/terms/
                  
                  
                    
                • Neil M.
                  I had some free time, something like this? http://www.nabber.org/media/HR3590.html Neil
                  Message 8 of 14 , May 25, 2010
                  • 0 Attachment
                    I had some free time, something like this?

                    http://www.nabber.org/media/HR3590.html

                    Neil

                    On 5/25/2010 7:23 PM, Harvey Frey wrote:
                    >
                    >
                    > Hi Neil:
                    >
                    > Thanks for the response!
                    >
                    > I simply want to use the 'contents' section of a bill to hyperlink
                    > to the actual text paragraphs within the same bill, to make the bills
                    > easier to navigate and comprehend.
                    >
                    > If I save the contents page from Thomas, the links point to a cgi
                    > script and disappear within a few minutes. AFAIK, Thomas uses no
                    > permanent links - everything runs through their cgi script.
                    >
                    > If I download the entire bill from Thomas, it contains no links at
                    > all, so I need to be able to find section headings and put name anchors
                    > there, and put corresponding href anchors in the correct TOC line, but
                    > not do it for incidental references which are not headings.
                    >
                    > So it's not a problem of finding stereotypical text references to
                    > USC sections and constructing HTML from them. That would be a pretty
                    > straightforward RegExp problem.
                    >
                    > Harvey
                    > ================================================
                    >
                    > Neil M. wrote:
                    >
                    >> I've got an old python script that takes the US Code text files (from
                    >> http://uscode.house.gov), scans for references and outputs mediawiki
                    >> formatted pages and wiki links. It shouldn't be too difficult to modify
                    >> it to parse some other text/HTML file and output HTML instead. What is
                    >> it you want to link to exactly? The aforementioned US Code website?
                    >> Thomas? Both?
                    >>
                    >> Neil
                    >>
                    >> On 5/24/2010 3:59 PM, Harvey Frey wrote:
                    >>
                    >>>
                    >>>
                    >>> Has anyone tried downloading bills from Thomas in HTML format, or
                    >>> converting their Text downloads to HTML?
                    >>>
                    >>> The guy at the Thomas help desk at first didn't understand what I
                    >>> wanted, and then said that no one had ever asked for that before.
                    >>>
                    >>> As you know, when you download the Contents page of a bill, the
                    >>> hyperlinks to the actual sections point to your own local folder instead
                    >>> of to the Thomas page where they exist.
                    >>>
                    >>> If you add a base statement to point them back to the Thomas site,
                    >>> the links disappear in a few minutes since the search expires.
                    >>>
                    >>> If you download all their referenced pages, you can't use their
                    >>> links to convert them to local hyperlinks, since they address them
                    >>> through a cgi program instead of through static name anchors.
                    >>>
                    >>> Your can download a bill as plain text rather than HTML, but
                    >>> manually adding all the name anchors and hyperlinks would be a massive job.
                    >>>
                    >>> (I did it for the Patriot Act, and it wasn't fun, especially since
                    >>> it looked like they intentionally obfuscated the references, sometimes
                    >>> using Public Law References, sometimes USC, and sometimes common names,
                    >>> so it was a detective job to find the sections they were amending.)
                    >>>
                    >>> Has anyone tried this, say with perl?
                    >>>
                    >>> I'm specifically interested in HR 3590, the recent Health Reform
                    >>> Bill. Some bills are posted as a massive XML file, but this one isn't.
                    >>> (If it were, I suppose you could use their id/id-ref pairs to construct
                    >>> href/name pairs, and then clean out the rest of the XML cruft.)
                    >>>
                    >>> Harvey
                    >>>
                    >>> =============================
                    >>> Harvey S. Frey MD PhD Esq.
                    >>> hsfrey@... www.harp.org
                    >>> -----------------------------
                    >>> "Withdrawing in disgust is not the same thing as apathy."
                    >>> - Brian Eno
                    >>> =============================
                    >>>
                    >>>
                    >>>
                    >>
                    >>
                    >> ------------------------------------
                    >>
                    >> Yahoo! Groups Links
                    >>
                    >>
                    >>
                    >>
                    >>
                    >
                  • Harvey Frey
                    Neil: Precisely! :-D Thank You! Did you write a script to do that? Manually it would surely have taken more than a little free time !! Harvey
                    Message 9 of 14 , May 26, 2010
                    • 0 Attachment
                      Neil:

                          Precisely! :-D
                          Thank You!

                          Did you write a script to do that?
                          Manually it would surely have taken more than a little "free time" !!

                      Harvey

                      Neil M. wrote:
                      I had some free time, something like this?
                      
                      http://www.nabber.org/media/HR3590.html
                      
                      Neil
                      
                      On 5/25/2010 7:23 PM, Harvey Frey wrote:
                        
                       
                      
                      Hi Neil:
                      
                          Thanks for the response!
                      
                          I simply want to use the 'contents' section of a bill to hyperlink
                      to the actual text paragraphs within the same bill, to make the bills
                      easier to navigate and comprehend.
                      
                          If I save the contents page from Thomas, the links point to a cgi
                      script and disappear within a few minutes. AFAIK, Thomas uses no
                      permanent links - everything runs through their cgi script.
                      
                          If I download the entire bill from Thomas, it contains no links at
                      all, so I need to be able to find section headings and put name anchors
                      there, and put corresponding href anchors in the correct TOC line, but
                      not do it for incidental references which are not headings.
                      
                          So it's not a problem of finding stereotypical text references to
                      USC sections and constructing HTML from them. That would be a pretty
                      straightforward RegExp problem.
                      
                      Harvey
                      ================================================
                      
                      Neil M. wrote:
                      
                          
                      I've got an old python script that takes the US Code text files (from
                      http://uscode.house.gov), scans for references and outputs mediawiki
                      formatted pages and wiki links.  It shouldn't be too difficult to modify
                      it to parse some other text/HTML file and output HTML instead.  What is
                      it you want to link to exactly?  The aforementioned US Code website?
                      Thomas?  Both?
                      
                      Neil
                      
                      On 5/24/2010 3:59 PM, Harvey Frey wrote:
                        
                            
                       
                      
                          Has anyone tried downloading bills from Thomas in HTML format, or
                      converting their Text downloads to HTML?
                      
                          The guy at the Thomas help desk at first didn't understand what I
                      wanted, and then said that no one had ever asked for that before.
                      
                          As you know, when you download the Contents page of a bill, the
                      hyperlinks to the actual sections point to your own local folder instead
                      of to the Thomas page where they exist.
                      
                          If you add a base statement to point them back to the Thomas site,
                      the links disappear in a few minutes since the search expires.
                      
                          If you download all their referenced pages, you can't use their
                      links to convert them to local hyperlinks, since they address them
                      through a cgi program instead of through static name anchors.
                      
                          Your can download a bill as plain text rather than HTML, but
                      manually adding all the name anchors and hyperlinks would be a massive job.
                      
                          (I did it for the Patriot Act, and it wasn't fun, especially since
                      it looked like they intentionally obfuscated the references, sometimes
                      using Public Law References, sometimes USC, and sometimes common names,
                      so it was a detective job to find the sections they were amending.)
                      
                          Has anyone tried this, say with perl?
                      
                          I'm specifically interested in HR 3590, the recent Health Reform
                      Bill. Some bills are posted as a massive XML file, but this one isn't.
                      (If it were, I suppose you could use their id/id-ref pairs to construct
                      href/name pairs, and then clean out the rest of the XML cruft.)
                      
                      Harvey
                      
                      =============================
                      Harvey S. Frey MD PhD Esq.
                      hsfrey@...  www.harp.org
                      -----------------------------
                      "Withdrawing in disgust is not the same thing as apathy."
                      - Brian Eno
                      =============================
                      
                      
                          
                              
                      ------------------------------------
                      
                      Yahoo! Groups Links
                      
                      
                      
                      
                        
                            
                      
                      ------------------------------------
                      
                      Yahoo! Groups Links
                      
                      <*> To visit your group on the web, go to:
                          http://groups.yahoo.com/group/govtrack/
                      
                      <*> Your email settings:
                          Individual Email | Traditional
                      
                      <*> To change settings online go to:
                          http://groups.yahoo.com/group/govtrack/join
                          (Yahoo! ID required)
                      
                      <*> To change settings via email:
                          govtrack-digest@yahoogroups.com 
                          govtrack-fullfeatured@yahoogroups.com
                      
                      <*> To unsubscribe from this group, send an email to:
                          govtrack-unsubscribe@yahoogroups.com
                      
                      <*> Your use of Yahoo! Groups is subject to:
                          http://docs.yahoo.com/info/terms/
                      
                      
                        
                    • Neil M.
                      Yes I wrote a quick Python script if anyone wants it just let me know. Neil
                      Message 10 of 14 , May 26, 2010
                      • 0 Attachment
                        Yes I wrote a quick Python script if anyone wants it just let me know.

                        Neil

                        On 5/26/2010 11:05 AM, Harvey Frey wrote:
                        >
                        >
                        > Neil:
                        >
                        > Precisely! :-D
                        > Thank You!
                        >
                        > Did you write a script to do that?
                        > Manually it would surely have taken more than a little "free time" !!
                        >
                        > Harvey
                        >
                        > Neil M. wrote:
                        >
                        >> I had some free time, something like this?
                        >>
                        >> http://www.nabber.org/media/HR3590.html
                        >>
                        >> Neil
                        >>
                        >> On 5/25/2010 7:23 PM, Harvey Frey wrote:
                        >>
                        >>>
                        >>>
                        >>> Hi Neil:
                        >>>
                        >>> Thanks for the response!
                        >>>
                        >>> I simply want to use the 'contents' section of a bill to hyperlink
                        >>> to the actual text paragraphs within the same bill, to make the bills
                        >>> easier to navigate and comprehend.
                        >>>
                        >>> If I save the contents page from Thomas, the links point to a cgi
                        >>> script and disappear within a few minutes. AFAIK, Thomas uses no
                        >>> permanent links - everything runs through their cgi script.
                        >>>
                        >>> If I download the entire bill from Thomas, it contains no links at
                        >>> all, so I need to be able to find section headings and put name anchors
                        >>> there, and put corresponding href anchors in the correct TOC line, but
                        >>> not do it for incidental references which are not headings.
                        >>>
                        >>> So it's not a problem of finding stereotypical text references to
                        >>> USC sections and constructing HTML from them. That would be a pretty
                        >>> straightforward RegExp problem.
                        >>>
                        >>> Harvey
                        >>> ================================================
                        >>>
                        >>> Neil M. wrote:
                        >>>
                        >>>
                        >>>> I've got an old python script that takes the US Code text files (from
                        >>>> http://uscode.house.gov), scans for references and outputs mediawiki
                        >>>> formatted pages and wiki links. It shouldn't be too difficult to modify
                        >>>> it to parse some other text/HTML file and output HTML instead. What is
                        >>>> it you want to link to exactly? The aforementioned US Code website?
                        >>>> Thomas? Both?
                        >>>>
                        >>>> Neil
                        >>>>
                        >>>> On 5/24/2010 3:59 PM, Harvey Frey wrote:
                        >>>>
                        >>>>
                        >>>>>
                        >>>>>
                        >>>>> Has anyone tried downloading bills from Thomas in HTML format, or
                        >>>>> converting their Text downloads to HTML?
                        >>>>>
                        >>>>> The guy at the Thomas help desk at first didn't understand what I
                        >>>>> wanted, and then said that no one had ever asked for that before.
                        >>>>>
                        >>>>> As you know, when you download the Contents page of a bill, the
                        >>>>> hyperlinks to the actual sections point to your own local folder instead
                        >>>>> of to the Thomas page where they exist.
                        >>>>>
                        >>>>> If you add a base statement to point them back to the Thomas site,
                        >>>>> the links disappear in a few minutes since the search expires.
                        >>>>>
                        >>>>> If you download all their referenced pages, you can't use their
                        >>>>> links to convert them to local hyperlinks, since they address them
                        >>>>> through a cgi program instead of through static name anchors.
                        >>>>>
                        >>>>> Your can download a bill as plain text rather than HTML, but
                        >>>>> manually adding all the name anchors and hyperlinks would be a massive job.
                        >>>>>
                        >>>>> (I did it for the Patriot Act, and it wasn't fun, especially since
                        >>>>> it looked like they intentionally obfuscated the references, sometimes
                        >>>>> using Public Law References, sometimes USC, and sometimes common names,
                        >>>>> so it was a detective job to find the sections they were amending.)
                        >>>>>
                        >>>>> Has anyone tried this, say with perl?
                        >>>>>
                        >>>>> I'm specifically interested in HR 3590, the recent Health Reform
                        >>>>> Bill. Some bills are posted as a massive XML file, but this one isn't.
                        >>>>> (If it were, I suppose you could use their id/id-ref pairs to construct
                        >>>>> href/name pairs, and then clean out the rest of the XML cruft.)
                        >>>>>
                        >>>>> Harvey
                        >>>>>
                        >>>>> =============================
                        >>>>> Harvey S. Frey MD PhD Esq.
                        >>>>> hsfrey@... www.harp.org
                        >>>>> -----------------------------
                        >>>>> "Withdrawing in disgust is not the same thing as apathy."
                        >>>>> - Brian Eno
                        >>>>> =============================
                        >>>>>
                        >>>>>
                        >>>>>
                        >>>>>
                        >>>> ------------------------------------
                        >>>>
                        >>>> Yahoo! Groups Links
                        >>>>
                        >>>>
                        >>>>
                        >>>>
                        >>>>
                        >>>>
                        >>
                        >>
                        >> ------------------------------------
                        >>
                        >> Yahoo! Groups Links
                        >>
                        >>
                        >>
                        >>
                        >>
                        >
                      • Harvey Frey
                        ... Yes, please! It does need a little manual post-editing, since the section numbers in amended texts can be (and are) duplicates. Harvey
                        Message 11 of 14 , May 26, 2010
                        • 0 Attachment
                          Neil:

                          >
                          if anyone wants it just let me know<

                               Yes, please!

                              It does need a little manual post-editing, since the section numbers in amended texts can be (and are) duplicates.

                          Harvey

                          Neil M. wrote:
                          Yes I wrote a quick Python script if anyone wants it just let me know.
                          
                          Neil
                          
                          On 5/26/2010 11:05 AM, Harvey Frey wrote:
                            
                           
                          
                          Neil:
                          
                              Precisely! :-D
                              Thank You!
                          
                              Did you write a script to do that?
                              Manually it would surely have taken more than a little "free time" !!
                          
                          Harvey
                          
                          Neil M. wrote:
                          
                              
                          I had some free time, something like this?
                          
                          http://www.nabber.org/media/HR3590.html
                          
                          Neil
                          
                          On 5/25/2010 7:23 PM, Harvey Frey wrote:
                            
                                
                           
                          
                          Hi Neil:
                          
                              Thanks for the response!
                          
                              I simply want to use the 'contents' section of a bill to hyperlink
                          to the actual text paragraphs within the same bill, to make the bills
                          easier to navigate and comprehend.
                          
                              If I save the contents page from Thomas, the links point to a cgi
                          script and disappear within a few minutes. AFAIK, Thomas uses no
                          permanent links - everything runs through their cgi script.
                          
                              If I download the entire bill from Thomas, it contains no links at
                          all, so I need to be able to find section headings and put name anchors
                          there, and put corresponding href anchors in the correct TOC line, but
                          not do it for incidental references which are not headings.
                          
                              So it's not a problem of finding stereotypical text references to
                          USC sections and constructing HTML from them. That would be a pretty
                          straightforward RegExp problem.
                          
                          Harvey
                          ================================================
                          
                          Neil M. wrote:
                          
                              
                                  
                          I've got an old python script that takes the US Code text files (from
                          http://uscode.house.gov), scans for references and outputs mediawiki
                          formatted pages and wiki links.  It shouldn't be too difficult to modify
                          it to parse some other text/HTML file and output HTML instead.  What is
                          it you want to link to exactly?  The aforementioned US Code website?
                          Thomas?  Both?
                          
                          Neil
                          
                          On 5/24/2010 3:59 PM, Harvey Frey wrote:
                            
                                
                                    
                           
                          
                              Has anyone tried downloading bills from Thomas in HTML format, or
                          converting their Text downloads to HTML?
                          
                              The guy at the Thomas help desk at first didn't understand what I
                          wanted, and then said that no one had ever asked for that before.
                          
                              As you know, when you download the Contents page of a bill, the
                          hyperlinks to the actual sections point to your own local folder instead
                          of to the Thomas page where they exist.
                          
                              If you add a base statement to point them back to the Thomas site,
                          the links disappear in a few minutes since the search expires.
                          
                              If you download all their referenced pages, you can't use their
                          links to convert them to local hyperlinks, since they address them
                          through a cgi program instead of through static name anchors.
                          
                              Your can download a bill as plain text rather than HTML, but
                          manually adding all the name anchors and hyperlinks would be a massive job.
                          
                              (I did it for the Patriot Act, and it wasn't fun, especially since
                          it looked like they intentionally obfuscated the references, sometimes
                          using Public Law References, sometimes USC, and sometimes common names,
                          so it was a detective job to find the sections they were amending.)
                          
                              Has anyone tried this, say with perl?
                          
                              I'm specifically interested in HR 3590, the recent Health Reform
                          Bill. Some bills are posted as a massive XML file, but this one isn't.
                          (If it were, I suppose you could use their id/id-ref pairs to construct
                          href/name pairs, and then clean out the rest of the XML cruft.)
                          
                          Harvey
                          
                          =============================
                          Harvey S. Frey MD PhD Esq.
                          hsfrey@...  www.harp.org
                          -----------------------------
                          "Withdrawing in disgust is not the same thing as apathy."
                          - Brian Eno
                          =============================
                          
                          
                              
                                  
                                      
                          ------------------------------------
                          
                          Yahoo! Groups Links
                          
                          
                          
                          
                            
                                
                                    
                          ------------------------------------
                          
                          Yahoo! Groups Links
                          
                          
                          
                          
                            
                                
                          
                          ------------------------------------
                          
                          Yahoo! Groups Links
                          
                          <*> To visit your group on the web, go to:
                              http://groups.yahoo.com/group/govtrack/
                          
                          <*> Your email settings:
                              Individual Email | Traditional
                          
                          <*> To change settings online go to:
                              http://groups.yahoo.com/group/govtrack/join
                              (Yahoo! ID required)
                          
                          <*> To change settings via email:
                              govtrack-digest@yahoogroups.com 
                              govtrack-fullfeatured@yahoogroups.com
                          
                          <*> To unsubscribe from this group, send an email to:
                              govtrack-unsubscribe@yahoogroups.com
                          
                          <*> Your use of Yahoo! Groups is subject to:
                              http://docs.yahoo.com/info/terms/
                          
                          
                            
                        Your message has been successfully submitted and would be delivered to recipients shortly.