projects / disclosr.git / commitdiff
search:
summary | shortlog | log | commit | commitdiff | tree
unified | plain
beginning of docx/pdf scrapers
beginning of docx/pdf scrapers


Former-commit-id: 72b18d2f2bae7cfce33fb8639ad1523c7bbcc0a3

file:a/admin/refreshDesignDoc.php -> file:b/admin/refreshDesignDoc.php
<?php <?php
   
require_once '../include/common.inc.php'; require_once '../include/common.inc.php';
//function createFOIDocumentsDesignDoc() { //function createFOIDocumentsDesignDoc() {
   
$foidb = $server->get_db('disclosr-foidocuments'); $foidb = $server->get_db('disclosr-foidocuments');
$obj = new stdClass(); $obj = new stdClass();
$obj->_id = "_design/" . urlencode("app"); $obj->_id = "_design/" . urlencode("app");
$obj->language = "javascript"; $obj->language = "javascript";
$obj->views->all->map = "function(doc) { emit(doc._id, doc); };"; $obj->views->all->map = "function(doc) { emit(doc._id, doc); };";
$obj->views->byDate->map = "function(doc) { emit(doc.date, doc); };"; $obj->views->byDate->map = "function(doc) { emit(doc.date, doc); };";
  $obj->views->byDate->reduce = "_count";
$obj->views->byDateMonthYear->map = "function(doc) { emit(doc.date, doc); };"; $obj->views->byDateMonthYear->map = "function(doc) { emit(doc.date, doc); };";
$obj->views->byDateMonthYear->reduce = "_count"; $obj->views->byDateMonthYear->reduce = "_count";
$obj->views->byAgencyID->map = "function(doc) { emit(doc.agencyID, doc); };"; $obj->views->byAgencyID->map = "function(doc) { emit(doc.agencyID, doc); };";
$obj->views->byAgencyID->reduce = "_count"; $obj->views->byAgencyID->reduce = "_count";
   
// allow safe updates (even if slightly slower due to extra: rev-detection check). // allow safe updates (even if slightly slower due to extra: rev-detection check).
$foidb->save($obj, true); $foidb->save($obj, true);
   
   
//function createDocumentsDesignDoc() { //function createDocumentsDesignDoc() {
$docdb = $server->get_db('disclosr-documents'); $docdb = $server->get_db('disclosr-documents');
   
$obj = new stdClass(); $obj = new stdClass();
$obj->_id = "_design/" . urlencode("app"); $obj->_id = "_design/" . urlencode("app");
$obj->language = "javascript"; $obj->language = "javascript";
$obj->views->web_server->map = "function(doc) {\n emit(doc.web_server, 1);\n}"; $obj->views->web_server->map = "function(doc) {\n emit(doc.web_server, 1);\n}";
$obj->views->web_server->reduce = "function (key, values, rereduce) {\n return sum(values);\n}"; $obj->views->web_server->reduce = "function (key, values, rereduce) {\n return sum(values);\n}";
$obj->views->byAgency->map = "function(doc) {\n emit(doc.agencyID, 1);\n}"; $obj->views->byAgency->map = "function(doc) {\n emit(doc.agencyID, 1);\n}";
$obj->views->byAgency->reduce = "function (key, values, rereduce) {\n return sum(values);\n}"; $obj->views->byAgency->reduce = "function (key, values, rereduce) {\n return sum(values);\n}";
$obj->views->byURL->map = "function(doc) {\n emit(doc.url, doc);\n}"; $obj->views->byURL->map = "function(doc) {\n emit(doc.url, doc);\n}";
$obj->views->agency->map = "function(doc) {\n emit(doc.agencyID, doc);\n}"; $obj->views->agency->map = "function(doc) {\n emit(doc.agencyID, doc);\n}";
$obj->views->byWebServer->map = "function(doc) {\n emit(doc.web_server, doc);\n}"; $obj->views->byWebServer->map = "function(doc) {\n emit(doc.web_server, doc);\n}";
$obj->views->getValidationRequired = "function(doc) {\nif (doc.mime_type == \"text/html\" \n&& typeof(doc.validation) == \"undefined\") {\n emit(doc._id, doc._attachments);\n}\n}"; $obj->views->getValidationRequired = "function(doc) {\nif (doc.mime_type == \"text/html\" \n&& typeof(doc.validation) == \"undefined\") {\n emit(doc._id, doc._attachments);\n}\n}";
   
   
   
   
//function createAgencyDesignDoc() { //function createAgencyDesignDoc() {
$db = $server->get_db('disclosr-agencies'); $db = $server->get_db('disclosr-agencies');
$obj = new stdClass(); $obj = new stdClass();
$obj->_id = "_design/" . urlencode("app"); $obj->_id = "_design/" . urlencode("app");
$obj->language = "javascript"; $obj->language = "javascript";
$obj->views->all->map = "function(doc) { emit(doc._id, doc); };"; $obj->views->all->map = "function(doc) { emit(doc._id, doc); };";
$obj->views->byABN->map = "function(doc) { emit(doc.abn, doc); };"; $obj->views->byABN->map = "function(doc) { emit(doc.abn, doc); };";
$obj->views->byCanonicalName->map = "function(doc) { $obj->views->byCanonicalName->map = "function(doc) {
if (doc.parentOrg || doc.orgType == 'FMA-DepartmentOfState') { if (doc.parentOrg || doc.orgType == 'FMA-DepartmentOfState') {
emit(doc.name, doc); emit(doc.name, doc);
} }
};"; };";
$obj->views->byDeptStateName->map = "function(doc) { $obj->views->byDeptStateName->map = "function(doc) {
if (doc.orgType == 'FMA-DepartmentOfState') { if (doc.orgType == 'FMA-DepartmentOfState') {
emit(doc.name, doc._id); emit(doc.name, doc._id);
} }
};"; };";
$obj->views->parentOrgs->map = "function(doc) { $obj->views->parentOrgs->map = "function(doc) {
if (doc.parentOrg) { if (doc.parentOrg) {
emit(doc._id, doc.parentOrg); emit(doc._id, doc.parentOrg);
} }
};"; };";
$obj->views->byName->map = 'function(doc) { $obj->views->byName->map = 'function(doc) {
if (typeof(doc["status"]) == "undefined" || doc["status"] != "suspended") { if (typeof(doc["status"]) == "undefined" || doc["status"] != "suspended") {
emit(doc.name, doc._id); emit(doc.name, doc._id);
if (typeof(doc.shortName) != "undefined" && doc.shortName != doc.name) { if (typeof(doc.shortName) != "undefined" && doc.shortName != doc.name) {
emit(doc.shortName, doc._id); emit(doc.shortName, doc._id);
} }
for (name in doc.otherNames) { for (name in doc.otherNames) {
if (doc.otherNames[name] != "" && doc.otherNames[name] != doc.name) { if (doc.otherNames[name] != "" && doc.otherNames[name] != doc.name) {
emit(doc.otherNames[name], doc._id); emit(doc.otherNames[name], doc._id);
} }
} }
for (name in doc.foiBodies) { for (name in doc.foiBodies) {
if (doc.foiBodies[name] != "" && doc.foiBodies[name] != doc.name) { if (doc.foiBodies[name] != "" && doc.foiBodies[name] != doc.name) {
emit(doc.foiBodies[name], doc._id); emit(doc.foiBodies[name], doc._id);
} }
} }
for (name in doc.positions) { for (name in doc.positions) {
if (doc.positions[name] != "" && doc.positions[name] != doc.name) { if (doc.positions[name] != "" && doc.positions[name] != doc.name) {
emit(doc.positions[name], doc._id); emit(doc.positions[name], doc._id);
} }
} }
} }
};'; };';
   
$obj->views->foiEmails->map = "function(doc) { $obj->views->foiEmails->map = "function(doc) {
emit(doc._id, doc.foiEmail); emit(doc._id, doc.foiEmail);
};"; };";
   
$obj->views->byLastModified->map = "function(doc) { emit(doc.metadata.lastModified, doc); }"; $obj->views->byLastModified->map = "function(doc) { emit(doc.metadata.lastModified, doc); }";
$obj->views->getActive->map = 'function(doc) { if (doc.status == "active") { emit(doc._id, doc); } };'; $obj->views->getActive->map = 'function(doc) { if (doc.status == "active") { emit(doc._id, doc); } };';
$obj->views->getSuspended->map = 'function(doc) { if (doc.status == "suspended") { emit(doc._id, doc); } };'; $obj->views->getSuspended->map = 'function(doc) { if (doc.status == "suspended") { emit(doc._id, doc); } };';
$obj->views->getScrapeRequired->map = "function(doc) { $obj->views->getScrapeRequired->map = "function(doc) {
   
var lastScrape = Date.parse(doc.metadata.lastScraped); var lastScrape = Date.parse(doc.metadata.lastScraped);
   
var today = new Date(); var today = new Date();
   
if (!lastScrape || lastScrape.getTime() + 1000 != today.getTime()) { if (!lastScrape || lastScrape.getTime() + 1000 != today.getTime()) {
emit(doc._id, doc); emit(doc._id, doc);
} }
   
};"; };";
$obj->views->showNamesABNs->map = "function(doc) { emit(doc._id, {name: doc.name, abn: doc.abn}); };"; $obj->views->showNamesABNs->map = "function(doc) { emit(doc._id, {name: doc.name, abn: doc.abn}); };";
$obj->views->getConflicts->map = "function(doc) { $obj->views->getConflicts->map = "function(doc) {
if (doc._conflicts) { if (doc._conflicts) {
emit(null, [doc._rev].concat(doc._conflicts)); emit(null, [doc._rev].concat(doc._conflicts));
} }
}"; }";
// http://stackoverflow.com/questions/646628/javascript-startswith // http://stackoverflow.com/questions/646628/javascript-startswith
$obj->views->score->map = 'if(!String.prototype.startsWith){ $obj->views->score->map = 'if(!String.prototype.startsWith){
String.prototype.startsWith = function (str) { String.prototype.startsWith = function (str) {
return !this.indexOf(str); return !this.indexOf(str);
} }
} }
   
function(doc) { function(doc) {
count = 0; count = 0;
if (doc["status"] != "suspended") { if (doc["status"] != "suspended") {
for(var propName in doc) { for(var propName in doc) {
if(typeof(doc[propName]) != "undefined" && doc[propName] != "") { if(typeof(doc[propName]) != "undefined" && doc[propName] != "") {
count++; count++;
} }
} }
portfolio = doc.parentOrg; portfolio = doc.parentOrg;
if (doc.orgType == "FMA-DepartmentOfState") { if (doc.orgType == "FMA-DepartmentOfState") {
portfolio = doc._id; portfolio = doc._id;
} }
if (doc.orgType == "Court-Commonwealth" || doc.orgType == "FMA-DepartmentOfParliament") { if (doc.orgType == "Court-Commonwealth" || doc.orgType == "FMA-DepartmentOfParliament") {
portfolio = doc.orgType; portfolio = doc.orgType;
} }
emit(count+doc._id, {id:doc._id, name: doc.name, score:count, orgType: doc.orgType, portfolio:portfolio}); emit(count+doc._id, {id:doc._id, name: doc.name, score:count, orgType: doc.orgType, portfolio:portfolio});
} }
}'; }';
$obj->views->scoreHas->map = 'if(!String.prototype.startsWith){ $obj->views->scoreHas->map = 'if(!String.prototype.startsWith){
String.prototype.startsWith = function (str) { String.prototype.startsWith = function (str) {
return !this.indexOf(str); return !this.indexOf(str);
} }
} }
if(!String.prototype.endsWith){ if(!String.prototype.endsWith){
String.prototype.endsWith = function(suffix) { String.prototype.endsWith = function(suffix) {
    return this.indexOf(suffix, this.length - suffix.length) !== -1;     return this.indexOf(suffix, this.length - suffix.length) !== -1;
}; };
} }
function(doc) { function(doc) {
if (typeof(doc["status"]) == "undefined" || doc["status"] != "suspended") { if (typeof(doc["status"]) == "undefined" || doc["status"] != "suspended") {
for(var propName in doc) { for(var propName in doc) {
if(typeof(doc[propName]) != "undefined" && (propName.startsWith("has") || propName.endsWith("URL"))) { if(typeof(doc[propName]) != "undefined" && (propName.startsWith("has") || propName.endsWith("URL"))) {
emit(propName, 1); emit(propName, 1);
} }
} }
emit("total", 1); emit("total", 1);
} }
}'; }';
$obj->views->scoreHas->reduce = 'function (key, values, rereduce) { $obj->views->scoreHas->reduce = 'function (key, values, rereduce) {
return sum(values); return sum(values);
}'; }';
$obj->views->fieldNames->map = ' $obj->views->fieldNames->map = '
function(doc) { function(doc) {
for(var propName in doc) { for(var propName in doc) {
emit(propName, doc._id); emit(propName, doc._id);
} }
}'; }';
$obj->views->fieldNames->reduce = 'function (key, values, rereduce) { $obj->views->fieldNames->reduce = 'function (key, values, rereduce) {
return values.length; return values.length;
}'; }';
// allow safe updates (even if slightly slower due to extra: rev-detection check). // allow safe updates (even if slightly slower due to extra: rev-detection check).
$db->save($obj, true); $db->save($obj, true);
?> ?>
   
file:a/documents/about.php (deleted)
<?php  
 
include('template.inc.php');  
include_header_documents("");  
include_once('../include/common.inc.php');  
?>  
<h1>About</h1>  
<?php  
include_footer_documents();  
?>  
 
file:a/documents/genericScrapers.py -> file:b/documents/genericScrapers.py
import sys,os import sys,os
sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../'))
import scrape import scrape
from bs4 import BeautifulSoup from bs4 import BeautifulSoup
from time import mktime from time import mktime
import feedparser import feedparser
import abc import abc
import unicodedata, re import unicodedata, re
import dateutil import dateutil
from dateutil.parser import * from dateutil.parser import *
from datetime import * from datetime import *
import codecs import codecs
   
  from StringIO import StringIO
   
  from docx import *
  from lxml import etree
  import zipfile
   
  from pdfminer.pdfparser import PDFDocument, PDFParser
  from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf
  from pdfminer.pdfdevice import PDFDevice, TagExtractor
  from pdfminer.converter import TextConverter
  from pdfminer.cmapdb import CMapDB
  from pdfminer.layout import LAParams
   
class GenericDisclogScraper(object): class GenericDisclogScraper(object):
__metaclass__ = abc.ABCMeta __metaclass__ = abc.ABCMeta
agencyID = None agencyID = None
disclogURL = None disclogURL = None
def remove_control_chars(self, input): def remove_control_chars(self, input):
return "".join([i for i in input if ord(i) in range(32, 127)]) return "".join([i for i in input if ord(i) in range(32, 127)])
def getAgencyID(self): def getAgencyID(self):
""" disclosr agency id """ """ disclosr agency id """
if self.agencyID == None: if self.agencyID == None:
self.agencyID = os.path.basename(sys.argv[0]).replace(".py","") self.agencyID = os.path.basename(sys.argv[0]).replace(".py","")
return self.agencyID return self.agencyID
   
def getURL(self): def getURL(self):
""" disclog URL""" """ disclog URL"""
if self.disclogURL == None: if self.disclogURL == None:
agency = scrape.agencydb.get(self.getAgencyID()) agency = scrape.agencydb.get(self.getAgencyID())
self.disclogURL = agency['FOIDocumentsURL'] self.disclogURL = agency['FOIDocumentsURL']
return self.disclogURL return self.disclogURL
   
@abc.abstractmethod @abc.abstractmethod
def doScrape(self): def doScrape(self):
""" do the scraping """ """ do the scraping """
return return
   
@abc.abstractmethod class GenericPDFDisclogScraper(GenericDisclogScraper):
def getDescription(self, content, entry, doc):  
""" get description""" def doScrape(self):
return foidocsdb = scrape.couch['disclosr-foidocuments']
  (url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID())
   
  laparams = LAParams()
   
  rsrcmgr = PDFResourceManager(caching=True)
   
  outfp = StringIO.StringIO()
   
  device = TextConverter(rsrcmgr, outfp, codec='utf-8', laparams=laparams)
   
   
  fp = StringIO.StringIO()
  fp.write(content)
  description = output.getvalue();
  process_pdf(rsrcmgr, device, fp, set(), caching=True, check_extractable=True)
  fp.close()
  device.close()
  outfp.close()
   
  hash = scrape.mkhash(description)
  #print hash
  doc = foidocsdb.get(hash)
  #print doc
  if doc == None:
  print "saving "+ hash
  edate = datetime.fromtimestamp(mktime( )).strftime("%Y-%m-%d")
  doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': self.getURL(), 'docID': hash,
  "date": edate,"title": "Disclosure Log Updated"}
  self.getDescription(entry,entry, doc)
  foidocsdb.save(doc)
  else:
  print "already saved"
   
   
  class GenericDOCXDisclogScraper(GenericDisclogScraper):
   
  def doScrape(self):
  foidocsdb = scrape.couch['disclosr-foidocuments']
  (url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID())
   
  mydoc = zipfile.ZipFile(file)
  xmlcontent = mydoc.read('word/document.xml')
  document = etree.fromstring(xmlcontent)
   
  ## Fetch all the text out of the document we just created
  paratextlist = getdocumenttext(document)
   
  # Make explicit unicode version
  newparatextlist = []
  for paratext in paratextlist:
  newparatextlist.append(paratext.encode("utf-8"))
   
  ## Print our documnts test with two newlines under each paragraph
  description = '\n\n'.join(newparatextlist)
   
  hash = scrape.mkhash(description)
  #print hash
  doc = foidocsdb.get(hash)
  #print doc
  if doc == None:
  print "saving "+ hash
  edate = datetime.fromtimestamp(mktime()).strftime("%Y-%m-%d")
  doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': self.getURL(), 'docID': hash,
  "date": edate,"title": "Disclosure Log Updated"}
  self.getDescription(entry,entry, doc)
  foidocsdb.save(doc)
  else:
  print "already saved"
   
   
class GenericRSSDisclogScraper(GenericDisclogScraper): class GenericRSSDisclogScraper(GenericDisclogScraper):
   
def doScrape(self): def doScrape(self):
foidocsdb = scrape.couch['disclosr-foidocuments'] foidocsdb = scrape.couch['disclosr-foidocuments']
(url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID()) (url,mime_type,content) = scrape.fetchURL(scrape.docsdb, self.getURL(), "foidocuments", self.getAgencyID())
feed = feedparser.parse(content) feed = feedparser.parse(content)
for entry in feed.entries: for entry in feed.entries:
#print entry #print entry
print entry.id print entry.id
hash = scrape.mkhash(entry.id) hash = scrape.mkhash(entry.id)
#print hash #print hash
doc = foidocsdb.get(hash) doc = foidocsdb.get(hash)
#print doc #print doc
if doc == None: if doc == None:
print "saving "+ hash print "saving "+ hash
edate = datetime.fromtimestamp(mktime( entry.published_parsed)).strftime("%Y-%m-%d") edate = datetime.fromtimestamp(mktime( entry.published_parsed)).strftime("%Y-%m-%d")
doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': entry.link, 'docID': entry.id, doc = {'_id': hash, 'agencyID': self.getAgencyID(), 'url': entry.link, 'docID':