Nsryjdtyk

This is built with Python 2.7.15.

The goal of this script is to count the number of words spoken by each Senator on the floor of Congress between given dates.

It pulls from the Congressional Record, cuts out the sections not spoken on the floor (eg lists of Amendments) and returns a count of words by senator. The current list of senators is from the 115th Congress, from 2017 to 2018

# -*- coding: utf-8 -*-

from requests import get  # to make GET request

import time

from datetime import date, timedelta

import os.path

import textract

import re

import csv

import sys



#download URLs

def downloadPDF(url, file_name):

    #get request

    response = get(url)

    content_type = response.headers.get('content-type')



    if content_type == "application/pdf":

        # open in binary mode

        save_path = '/Users/One/Document/Workspace/Projects/Senate/Congressional_Record/'

        path = os.path.join(save_path, file_name)



        with open(path, "wb") as file:

            #write to file

            file.write(response.content)

            return 'SUCCESS'



#download congressional records and extract text

def downloadConRecords(startDate, endDate): #YYYYMMDD

    #initialize lists

    congressionalRecords = 

    dates = 



    #convert args to date-type

    startDate = str(startDate)

    endDate = str(endDate)

    startDate = date(int(startDate[0:4]),int(startDate[4:6]),int(startDate[6:8]))

    endDate = date(int(endDate[0:4]),int(endDate[4:6]),int(endDate[6:8]))



    #list all the days to check

    range = endDate - startDate

    i=0

    while i <= range.days:

        dates += [str(startDate + timedelta(days=i))]

        i+=1



    #try URL, download PDF, add to list object

    for i in dates:

        YYYY = i[0:4]

        MM = i[5:7]

        DD = i[8:10]

        url = ('https://www.congress.gov/crec/' + YYYY + '/'+ MM + '/' + DD + '/CREC-' + YYYY + '-' + MM + '-' + DD + '-senate.pdf')

        filename = (YYYY+MM+DD+".pdf")



        time.sleep(20)



        if downloadPDF(url,filename) == "SUCCESS":

            print(url)

            textExtract = 

            textExtract = textract.process('/Users/One/Document/Workspace/Projects/Senate/Congressional_Record/' + filename)

            #makes a list with the filename and the text

            entry = [url, filename[:-4].strip(), textExtract]

            congressionalRecords += [entry]



    return congressionalRecords



#process the congressional records

def cleanRecords(congressionalRecords, phraseDelete, segmentStart, segmentEnd):

    #eliminate n/, phrases, segments

    for record in congressionalRecords:

        clean = 

        dump = 

        clean = record[2].split('n')

        clean = " ".join(clean)



        for i in phraseDelete:

            clean = clean.replace(i,'')



        for start in enumerate(segmentStart):

            erase = re.findall('{}.*?{}'.format(start[1], segmentEnd), clean)

            clean = re.sub('{}.*?{}'.format(start[1], segmentEnd), ' ', clean)

            dump += [erase]



        record += [clean]

        record += [dump]



    return congressionalRecords



#split the text into sections based on the speaker

def multiDelimStringSplitter(aString, separators):

    # separators is an array of strings that are being used to split the the string.

    # sort separators in order of descending length

    separators.sort(key=len)

    listToReturn = 

    rows = 

    l = 

    i = 0



    while i < len(aString):

        theSeparator = ""

        for current in separators:

            if current == aString[i:i+len(current)]:

                #if this section is the separateor, then set the variable to that separator

                theSeparator = current

        if theSeparator != "":

            listToReturn += [theSeparator]

            i = i + len(theSeparator)

        else:

            if listToReturn == :

                listToReturn = [""]

            if(listToReturn[-1] in separators):

                listToReturn += [""]

            #adds the text one character at a time

            listToReturn[-1] += aString[i]

            i += 1



    i = 0

    listToReturn = listToReturn[1:len(listToReturn)]



    while (i+1) < len(listToReturn):

        #initialize entry with speaker and text

        entry = listToReturn[i:i+2]

        #get word count and add it to the entry

        text = listToReturn[i+1]

        #strip punctuation out of the string

        text = re.sub(r'[^ws]','',text)

        wordcount = str(len(text.split()))

        entry += [wordcount]

        #add the entry as its own list

        rows.append(entry)

        i += 2



    return rows



#get information on the records, speakers and word counts

def writeAnalytics(congressionalRecords, senators):

    textExport = [['url','date','extractRaw','extractClean']]

    textDump = [['date','extractDump']]

    speechExport = [['date','speaker','text','wordCount']]

    speakerExport = [['date','speaker','wordCount']]

    speakerAnalytics = [['startDate', 'endDate', 'speaker', 'wordCount']]

    perSpeakerTotal = 

    recordDates = 



    for i in senators:

        # wordsperSpeech += [i,0]

        perSpeakerTotal += [i,0]



    #add delimiter separated text extracts

    for record in congressionalRecords:

        splitLines = 

        wordsperSpeaker = 



        #initialize the lists

        for i in senators:

            # wordsperSpeech += [i,0]

            wordsperSpeaker += [i,0]



        #split the extracts into each speech

        splitLines = multiDelimStringSplitter(record[3],senators)

        record += [splitLines]



        #prep the dump export

        for i in record[4]:

            if len(i) > 0:

                textDump += [[record[1], i]]



        #add the words per speaker

        for speech in record[5]:

            speechExport += [[record[1],speech[0],speech[1],speech[2]]]

            #find the speaker -> pull that person's wordsperSpeaker and add the speech's word count

            wordsperSpeaker[wordsperSpeaker.index(speech[0])+1] += int(speech[2])



        record += [wordsperSpeaker]

        textExport += [[record[0],record[1],record[2],record[3]]]

        recordDates += [record[1]]



        #prepare export for the words per speaker

        for i in senators:

            speakerExport += [[record[1],i, wordsperSpeaker[wordsperSpeaker.index(i)+1]]]

            #add word counts for senators who spoke

            #find senator's index in perSpeaker total -> add the words spoken in that session (record)

            perSpeakerTotal[perSpeakerTotal.index(i)+1] += record[6][record[6].index(i)+1]



    for i in senators:

        speakerAnalytics += [[min(recordDates), max(recordDates), i, perSpeakerTotal[perSpeakerTotal.index(i)+1]]]



    return textExport, textDump, speechExport, speakerExport, speakerAnalytics



#export to CSV

def exportCSV(flatList, filename):

    with open(filename+".csv", "wb") as f:

        writer = csv.writer(f)

        writer.writerows(flatList)



phraseDelete = [

',',

'This ‘‘bullet’’ symbol identifies statements or insertions which are not spoken by a Member of the Senate on the floor.'

]



segmentStart = [

' PRAYER ',

' PLEDGE OF ALLEGIANCE ',

' APPOINTMENT OF ACTING PRESIDENT PRO TEMPORE ',

' CERTIFICATES OF ELECTION ',

' MESSAGE FROM THE HOUSE ',

' MESSAGES FROM THE PRESIDENT ',

' EXECUTIVE MESSAGES REFERRED ',

' LIST OF SENATORS BY STATES ',

' PRESIDENTIAL MESSAGES ',

' ENROLLED BILLS PRESENTED ',

' MEASURES REFERRED ',

' EXECUTIVE AND OTHER COMMUNICATIONS ',

' REPORTS OF COMMITTEES ',

' EXECUTIVE REPORTS OF COMMITTEES ',

' AMENDMENTS SUBMITTED AND PROPOSED ',

' INTRODUCTION OF BILLS AND JOINT RESOLUTIONS ',

' ADDITIONAL COSPONSORS ',

' SUBMITTED RESOLUTIONS ',

' SUBMISSION OF CONCURRENT AND SENATE RESOLUTIONS ',

' SENATE RESOLUTION ',

' TEXT OF AMENDMENTS ',

' APPOINTMENT ',

' AUTHORITY FOR COMMITTEES TO MEET ',

' CONFIRMATION ',

' NOMINATION '

]



segmentEnd = ' f '



senators = ['Mr. SESSIONS', 'Mr. STRANGE', 'Mr. JONES', 'Mr. SHELBY', 'Mr. SULLIVAN',  'Ms. MURKOWSKI',

'Mr. FLAKE', 'Mr. MCCAIN', 'Mr. KYL', 'Mr. COTTON', 'Mr. BOOZMAN', 'Mrs. FEINSTEIN', 'Ms. HARRIS',

'Mr. GARDNER', 'Mr. BENNET', 'Mr. MURPHY', 'Mr. BLUMENTHAL', 'Mr. CARPER', 'Mr. COONS', 'Mr. NELSON', 'Mr. RUBIO',

'Mr. PERDUE', 'Mr. ISAKSON', 'Ms. HIRONO', 'Mr. SCHATZ', 'Mr. RISCH', 'Mr. CRAPO', 'Mr. DURBIN',

'Ms. DUCKWORTH', 'Mr. DONNELLY', 'Mr. YOUNG', 'Mrs. ERNST', 'Mr. GRASSLEY', 'Mr. ROBERTS', 'Mr. MORAN',

'Mr. MCCONNELL', 'Mr. PAUL', 'Mr. CASSIDY', 'Mr. KENNEDY', 'Mr. KING', 'Ms. COLLINS', 'Mr. CARDIN',

'Mr. VAN HOLLEN', 'Ms. WARREN', 'Mr. MARKEY', 'Ms. STABENOW', 'Mr. PETERS', 'Ms. KLOBUCHAR',

'Mr. FRANKEN', 'Ms. SMITH', 'Mr. WICKER', 'Mr. COCHRAN', 'Mrs. HYDE-SMITH',  'Mrs. MCCASKILL',

'Mr. BLUNT', 'Mr. TESTER', 'Mr. DAINES', 'Mrs. FISCHER', 'Mr. SASSE', 'Mr. HELLER', 'Ms. CORTEZ MASTO',

'Mrs. SHAHEEN', 'Ms. HASSAN',  'Mr. MENENDEZ', 'Mr. BOOKER', 'Mr. HEINRICH', 'Mr. UDALL',

'Mrs. GILLIBRAND', 'Mr. SCHUMER', 'Mr. TILLIS', 'Mr. BURR', 'Ms. HEITKAMP', 'Mr. HOEVEN',

'Mr. BROWN', 'Mr. PORTMAN', 'Mr. INHOFE', 'Mr. LANKFORD', 'Mr. MERKLEY', 'Mr. WYDEN',

'Mr. CASEY', 'Mr. TOOMEY', 'Mr. WHITEHOUSE', 'Mr. REED', 'Mr. GRAHAM',

'Mr. SCOTT', 'Mr. ROUNDS', 'Mr. THUNE', 'Mr. CORKER', 'Mr. ALEXANDER', 'Mr. CRUZ',

'Mr. CORNYN', 'Mr. HATCH', 'Mr. LEE', 'Mr. SANDERS', 'Mr. LEAHY', 'Mr. KAINE',

'Mr. WARNER', 'Ms. CANTWELL', 'Mrs. MURRAY', 'Mr. MANCHIN', 'Mrs. CAPITO',

'Ms. BALDWIN', 'Mr. JOHNSON', 'Mr. BARRASSO', 'Mr. ENZI',

'The ACTING PRESIDENT', 'The PRESIDING OFFICER', 'The VICE PRESIDENT'

'Executive nominations confirmed by']



startDate = '20170106'

endDate = '20170130'

conRecords = downloadConRecords(startDate, endDate)

cleanConRecords = cleanRecords(conRecords, phraseDelete, segmentStart, segmentEnd)

textExport, textDump, speechExport, speakerExport, speakerAnalytics = writeAnalytics(conRecords, senators)



exportCSV(textExport,'textExport')

exportCSV(textDump,'textDump')

exportCSV(speechExport,'speechExport')

exportCSV(speakerExport,'speakerExport')

exportCSV(speakerAnalytics,'speakerAnalytics')