release version 0.4.9

Readme.md

@@ -1,6 +1,6 @@
 # pdfgrab
 
-* Version 0.4.8-Pre
+* Version 0.4.9
 
 ## What is it?
 
@@ -9,21 +9,18 @@ Basically it analyses PDF files for Metadata. You can direct it to a file or dir
 You can show it the url of a pdf or use the integrated googlesearch (thanx to mario vilas class)
 to search for pdfs at target site, download and analyse them.
 
-## What is new in 0.4.8 bug fix pre-release?
+## What is new in 0.4.9?
 
-* catching google error at too many requests
-* catching dns resolve urlopen error at googlelib
-* fixing annoying bug in regard of pdfs behind urls like http://host/pdf/
-* fixing zero size pdf error(online linked pdfs which are not accessable)
-* added some logging
-
-## What is new in 0.4.7 release?
-
-* Added support for html output file, this will be placed in the outdir path and is more clear then a text or json file
-* Added basic logging support, logfile is placed in pdfgrab.py directory
-* Reordered Codebase, exported functionality to some libraries
-* PDF XMP Metadata is grabbed now as well, but not yet saved in output files
-* added docs/ section with Changelog and Todo
+* exported reporting methods to libreport.py
+* added optargs for disabling different report methods
+* made the html report a bit more shiny
+* added function for generating html report after analysis
+* exported requests and storing data to new library
+* code fixes and more clear error handling
+* removed necessary site: parameter at search flag -s
+* updated readme
+* -s flag now acceppts several domains
+* console logging more clean
 
 ## What information can be gathered?
 
@@ -132,7 +129,7 @@ Will analyse all pdf's in that directory
 
 ### Google Search Mode
 ```
-# ./pdfgrab.py -s site:kernel.org
+# ./pdfgrab.py -s kernel.org
 ```
 Result:
 ```
@@ -164,6 +161,26 @@ File: pdfgrab/bpf_global_data_and_static_keys.pdf
 /PTEX.Fullbanner This is pdfTeX, Version 3.14159265-2.6-1.40.17 (TeX Live 2016) kpathsea version 6.2.2
 ```
 
+### Google Search Mode, several domains
+```
+# ./pdfgrab.py -s example.com,example.us
+```
+
+### Reporting
+
+pdfgrab outputs the information in different formats. If not disabled by one of the reporting flags (see -h) you will
+find in the output directory:
+
+* html report
+* text report
+* text url list
+* json data
+* json url list
+
+### Logging
+
+pdfgrab creates a logfile in the running directory called "pdfgrab.log"
+
 ## Google
 
 * Search: filetype:pdf site:com

docs/Changelog

@@ -1,6 +1,20 @@
 Changelog
 =========
 
+Version 4.9
+-----------
+
+* exported reporting methods to libreport.py
+* added optargs for disabling different report methods
+* made the html report a bit more shiny
+* added function for generating html report after analysis
+* exported requests and storing data to new library
+* code fixes and more clear error handling
+* removed necessary site: parameter at search flag -s
+* updated readme
+* -s flag now acceppts several domains
+* console logging more clean
+
 Version 4.8 Bugfix-PreRelease
 -----------------------------
 

libs/libgoogle.py

@@ -5,19 +5,44 @@ from libs.libhelper import *
 def get_random_agent():
     return (gs.get_random_user_agent())
 
-def search_pdf(search, args):
+def hits_google(search, args):
+    ''' the function where googlesearch from mario vilas
+		is called
+	'''
+    s = search.split(',')
+    query = 'filetype:pdf'
+
+
+    try:
+        hits = gs.hits(query, domains=s,user_agent=gs.get_random_user_agent())
+
+    except urllib.error.HTTPError as e:
+        return False,e
+
+    except urllib.error.URLError as e:
+        return False,e
+
+    except IndexError as e:
+        return False,e
+
+    return True,hits
+
+
+def search_google(search, args):
     ''' the function where googlesearch from mario vilas
 		is called
 	'''
 
+    s = search.split(',')
     search_stop = args.search_stop
 
-    query = '%s filetype:pdf' % search
+    query = 'filetype:pdf'
+    #query = 'site:%s filetype:pdf' % search
     # print(query)
     urls = []
 
     try:
-        for url in gs.search(query, num=20, stop=search_stop, user_agent=gs.get_random_user_agent()):
+        for url in gs.search(query, num=20, domains=s,stop=search_stop, user_agent=gs.get_random_user_agent()):
             #print(url)
             urls.append(url)
 

libs/liblog.py

@@ -8,10 +8,12 @@ file_handler = logging.FileHandler('pdfgrab.log')
 console_handler = logging.StreamHandler()
 console_handler.setLevel(logging.WARNING)
 
-formatter = logging.Formatter('%(asctime)s:%(name)s:%(levelname)s:%(message)s')
+file_formatter = logging.Formatter('%(asctime)s:%(name)s:%(levelname)s:%(message)s')
+console_formatter = logging.Formatter('%(levelname)s:%(message)s')
 
-file_handler.setFormatter(formatter)
-console_handler.setFormatter(formatter)
+
+file_handler.setFormatter(file_formatter)
+console_handler.setFormatter(console_formatter)
 
 logger.addHandler(file_handler)
 logger.addHandler(console_handler)

libs/libreport.py

@@ -0,0 +1,173 @@
+import os
+import sys
+import json
+from json2html import * 
+from libs.pdf_png import get_png_base64
+
+def prepare_analysis_dict(ana_queue):
+    '''params: ana_queue - queue with collected information
+    '''
+    # initiate analysis dictionary
+    analysis_dict = {}
+
+    # move analysis dictionary in queue back to dictionary
+    while ana_queue.empty() == False:
+        item = ana_queue.get()
+        # print('item ', item)
+        analysis_dict.update(item)
+
+    # ana_q is empty now return the newly created dictionary
+    return analysis_dict
+
+def create_txt_report(analysis_dict, outdir, out_filename):
+    ''' create a txt report in the output directory
+    '''
+
+    # draw seperator lines
+    sep = '-' * 80 + '\n'
+
+    # create output filepath
+    txtout = "%s/%s.txt" % (outdir, out_filename)
+
+    # open the file and return filedescriptor
+    fwtxt = open(txtout, 'w')
+
+    # get the keys of the dict
+    for k in analysis_dict.keys():
+        # write seperator
+        fwtxt.write(sep)
+
+        # build entry filename of the pdf
+        fname = 'File: %s\n' % (analysis_dict[k]['filename'])
+
+        # build data entry
+        ddata = analysis_dict[k]['data']
+
+        # write the filename
+        fwtxt.write(fname)
+
+        # write the metadata
+        for kdata in ddata.keys():
+            metatxt = '%s:%s\n' % (kdata, ddata[kdata])
+            fwtxt.write(metatxt)
+
+        # write seperator
+        fwtxt.write(sep)
+
+    # close the file
+    fwtxt.close()
+
+    return True
+
+def create_json_report(analysis_dict, outdir, out_filename):
+    ''' create a jsonfile report in the output directory
+    '''
+
+    # build json output name
+    jsonout = "%s/%s.json" % (outdir, out_filename)
+
+    # open up json output file
+    fwjson = open(jsonout, 'w')
+
+    # convert dictionary to json data
+    jdata = json.dumps(analysis_dict)
+
+    # write json data to file 
+    fwjson.write(jdata)
+
+    # close file
+    fwjson.close()
+
+    return True
+
+def create_html_report(analysis_dict, outdir, out_filename):
+    ''' create a html report from json file using json2html in the output directory
+    '''
+
+    # build up path for html output file
+    htmlout = "%s/%s.html" % (outdir, out_filename)
+
+    # open htmlout filedescriptor
+    fwhtml = open(htmlout,'w')
+
+    # some html stuff
+    pdfpng=get_png_base64('supply/pdf_base64.png')
+    html_style ='<style>.center { display: block; margin-left: auto;margin-right: auto;} table {border-collapse: collapse;} th, td { border: 1px solid black;text-align: left; }</style>\n'
+    html_head = '<html><head><title>pdfgrab - {0} item/s</title>{1}</head>\n'.format(len(analysis_dict),html_style)
+    html_pdf_png = '<p class="center"><img class="center" src="data:image/jpeg;base64,{0}"><br><center>pdfgrab - grab and analyse pdf files</center><br></p>'.format(pdfpng)
+    html_body = '<body>{0}\n'.format(html_pdf_png)
+    html_end = '\n<br><br><p align="center"><a href="https://github.com/c0decave/pdfgrab">pdfgrab</a> by <a href="https://twitter.com/User_to_Root">dash</a></p></body></html>\n'
+
+    # some attributes
+    attr = 'id="meta-data" class="table table-bordered table-hover", border=1, cellpadding=3 summary="Metadata"'
+
+    # convert dictionary to json data
+    # in this mode each finding gets its own table there are other possibilities
+    # but now i go with this
+    html_out = ''
+    for k in analysis_dict.keys():
+        trans = analysis_dict[k]
+        jdata = json.dumps(trans)
+        html = json2html.convert(json = jdata, table_attributes=attr)
+        html_out = html_out + html + "\n"
+        #html_out = html_out + "<p>" + html + "</p>\n"
+    #jdata = json.dumps(analysis_dict)
+
+    # create html
+    #html = json2html.convert(json = jdata, table_attributes=attr)
+
+    # write html
+    fwhtml.write(html_head)
+    fwhtml.write(html_body)
+    fwhtml.write(html_out)
+    fwhtml.write(html_end)
+
+    # close html file
+    fwhtml.close()
+    
+def create_url_json(url_d, outdir, out_filename):
+    ''' create a json url file in output directory
+    '''
+
+    # create url savefile
+    jsonurlout = "%s/%s_url.json" % (outdir, out_filename)
+
+    # open up file for writting urls down
+    fwjson = open(jsonurlout, 'w')
+
+    # convert url dictionary to json
+    jdata = json.dumps(url_d)
+
+    # write json data to file
+    fwjson.write(jdata)
+
+    # close filedescriptor
+    fwjson.close()
+
+    return True
+
+def create_url_txt(url_d, outdir, out_filename):
+    ''' create a txt url file in output directory
+    '''
+    # build up txt out path
+    txtout = "%s/%s_url.txt" % (outdir, out_filename)
+
+    # open up our url txtfile
+    fwtxt = open(txtout, 'w')
+
+    # iterating through the keys of the url dictionary
+    for k in url_d.keys():
+
+        # get the entry
+        ddata = url_d[k]
+
+        # create meta data for saving
+        metatxt = '%s:%s\n' % (ddata['url'], ddata['filename'])
+
+        # write metadata to file
+        fwtxt.write(metatxt)
+
+    # close fd
+    fwtxt.close()
+
+    return True

libs/librequest.py

@@ -0,0 +1,162 @@
+import os
+import sys
+import json
+import socket
+import requests
+
+from libs.liblog import logger
+from libs.libhelper import *
+from libs.libgoogle import get_random_agent
+
+def store_file(url, data, outdir):
+    ''' storing the downloaded data to a file
+        params: url     - is used to create the filename
+                data    - the data of the file
+                outdir  - to store in which directory
+                returns: dict { "code":<code>, "data":<savepath>,"error":<error>} - the status code, the savepath, the errorcode
+    '''
+
+    logger.info('Store file {0}'.format(url))
+    name = find_name(url)
+
+    # only allow stored file a name with 50 chars
+    if len(name) > 50:
+        name = name[:49]
+
+    # build up the save path
+    save = "%s/%s" % (outdir, name)
+
+    try:
+        f = open(save, "wb")
+
+    except OSError as e:
+        logger.warning('store_file {0}'.format(e))
+        # return ret_dict
+        return {"code":False,"data":save,"error":e}
+
+    # write the data and return the written bytes
+    ret = f.write(data)
+
+    # check if bytes are zero
+    if ret == 0:
+        logger.warning('Written {0} bytes for file: {1}'.format(ret,save))
+
+    else:
+        # log to info that bytes and file has been written
+        logger.info('Written {0} bytes for file: {1}'.format(ret,save))
+
+    # close file descriptor
+    f.close()
+
+    # return ret_dict
+    return {"code":True,"data":save,"error":False}
+
+
+def download_file(url, args, header_data):
+    ''' downloading the file for later analysis 
+        params: url         - the url
+                args        - argparse args namespace
+                header_data - pre-defined header data
+        returns: ret_dict
+    '''
+
+    # check the remote tls certificate or not?
+    cert_check = args.cert_check
+
+    # run our try catch routine
+    try:
+        # request the url and save the response in req
+        # give header data and set verify as delivered by args.cert_check
+        req = requests.get(url, headers=header_data, verify=cert_check)
+
+    except requests.exceptions.SSLError as e:
+        logger.warning('download file {0}{1}'.format(url,e))
+
+        # return retdict
+        return {"code":False,"data":req,"error":e}
+
+    except requests.exceptions.InvalidSchema as e:
+        logger.warning('download file {0}{1}'.format(url,e))
+
+        # return retdict
+        return {"code":False,"data":False,"error":e}
+
+    except socket.gaierror as e:
+        logger.warning('download file, host not known {0} {1}'.format(url,e))
+        return {"code":False,"data":False,"error":e}
+
+    except:
+        logger.warning('download file, something wrong with remote server? {0}'.format(url))
+        # return retdict
+        if not req in locals():
+            req = False
+
+        return {"code":False,"data":req,"error":True}
+
+    #finally:
+        # lets close the socket
+        #req.close()
+
+    # return retdict
+    return {"code":True,"data":req,"error":False}
+
+def grab_run(url, args, outdir):
+    ''' function keeping all the steps for the user call of grabbing
+	just one and analysing it
+    '''
+    header_data = {'User-Agent': get_random_agent()}
+    rd_download = download_file(url, args, header_data)
+    code_down = rd_download['code']
+
+    # is code True download of file was successfull
+    if code_down:
+        rd_evaluate = evaluate_response(rd_download)
+        code_eval = rd_evaluate['code']
+        # if code is True, evaluation was also successful
+        if code_eval:
+            # get the content from the evaluate dictionary request
+            content = rd_evaluate['data'].content
+
+            # call store file 
+            rd_store = store_file(url, content, outdir)
+
+            # get the code
+            code_store = rd_store['code']
+
+            # get the savepath
+            savepath = rd_store['data']
+
+            # if code is True, storing of file was also successfull
+            if code_store:
+                return {"code":True,"data":savepath,"error":False}
+
+    return {"code":False,"data":False,"error":True}
+
+def evalute_content(ret_dict):
+    pass
+
+def evaluate_response(ret_dict):
+    ''' this method comes usually after download_file,
+        it will evaluate what has happened and if we even have some data to process
+        or not
+        params: data    - is the req object from the conducted request
+        return: {}
+        returns: dict { "code":<code>, "data":<savepath>,"error":<error>} - the status code, the savepath, the errorcode
+        '''
+    # extract data from ret_dict
+    req = ret_dict['data']
+
+    # get status code
+    url = req.url
+    status = req.status_code
+    reason = req.reason
+
+    # ahh everything is fine 
+    if status == 200:
+        logger.info('download file, {0} {1} {2}'.format(url,reason,status))
+        return {"code":True,"data":req,"error":False}
+
+    # nah something is not like it should be
+    else:
+        logger.warning('download file, {0} {1} {2}'.format(url,reason,status))
+        return {"code":False,"data":req,"error":True}

libs/pdf_png.py

@@ -0,0 +1,5 @@
+
+def get_png_base64(filename):
+    fr = open(filename,'r')
+    buf = fr.read()
+    return buf

pdfgrab.py

@@ -22,12 +22,14 @@ from PyPDF2 import pdf
 from libs.liblog import logger
 from libs.libhelper import *
 from libs.libgoogle import *
+from libs.libreport import *
+from libs.librequest import grab_run
 
 from IPython import embed
 
 # some variables in regard of the tool itself
 name = 'pdfgrab'
-version = '0.4.8-Pre'
+version = '0.4.9'
 author = 'dash'
 date = 'November 2019'
 
@@ -243,72 +245,6 @@ def check_encryption(filename):
 
     return True
 
-
-def download_pdf(url, args, header_data):
-    ''' downloading the pdfile for later analysis '''
-
-    # check the remote tls certificate or not?
-    cert_check = args.cert_check
-
-    try:
-        req = requests.get(url, headers=header_data, verify=cert_check)
-        # req = requests.get(url,headers=header_data,verify=False)
-        data = req.content
-        status_code = req.status_code
-
-    except requests.exceptions.SSLError as e:
-        logger.warning('download pdf {0}{1}'.format(url,e))
-        return -1
-
-    except:
-        logger.warning('download pdf, something wrong with remote server? {0}'.format(url))
-        return -1
-
-    if status_code == 403:
-        logger.warning('download pdf, 403 Forbidden {0}'.format(url))
-        return -1
-
-    # print(len(data))
-    return data
-
-
-def store_pdf(url, data, outdir):
-    ''' storing the downloaded pdf data
-    '''
-
-    logger.info('Store pdf {0}'.format(url))
-    name = find_name(url)
-    #logger.warning(url)
-    #logger.warning(name)
-    #logger.warning(outdir)
-
-    # only allow stored file a name with 50 chars
-    if len(name) > 50:
-        name = name[:49] + '.pdf'
-    # print(len(name))
-
-    save = "%s/%s" % (outdir, name)
-
-    try:
-        f = open(save, "wb")
-    except OSError as e:
-        logger.warning('store_pdf {0}'.format(e))
-        return -1
-
-    ret = f.write(data)
-    logger.info('Written {0} bytes for file: {1}'.format(ret,save))
-    f.close()
-
-    if ret == 0:
-        logger.warning('Written {0} bytes for file: {1}'.format(ret,save))
-        return save
-        #return -1
-
-
-    # return the savepath
-    return save
-
-
 def _parse_pdf(filename):
     ''' the real parsing function '''
 
@@ -320,26 +256,18 @@ def _parse_pdf(filename):
         logger.warning('Filesize is 0 bytes at file: {0}'.format(filename))
         return False
 
-
-def grab_url(url, args, outdir):
-    ''' function keeping all the steps for the user call of grabbing
-	just one pdf and analysing it
-    '''
-    header_data = {'User-Agent': get_random_agent()}
-    data = download_pdf(url, args, header_data)
-    if data != -1:
-        savepath = store_pdf(url, data, outdir)
-        _parse_pdf(savepath)
-
-    return
-
-
 def seek_and_analyse(search, args, outdir):
     ''' function for keeping all the steps of searching for pdfs and analysing
         them together
     '''
+    # check how many hits we got
+    # seems like the method is broken in googlsearch library :(
+    #code, hits = hits_google(search,args)
+    #if code:
+    #    print('Got {0} hits'.format(hits))
+
     # use the search function of googlesearch to get the results
-    code, values=search_pdf(search, args)
+    code, values=search_google(search, args)
     if not code:
         if values.code == 429:
             logger.warning('[-] Too many requests, time to change ip address or use proxychains')
@@ -362,7 +290,11 @@ def seek_and_analyse(search, args, outdir):
             item = url_q.get()
             # print(item)
             url = item['url']
-            grab_url(url, args, outdir)
+            rd_grabrun = grab_run(url, args, outdir)
+            code = rd_grabrun['code']
+            savepath = rd_grabrun['data']
+            if code:
+                _parse_pdf(savepath)
 
     return True
 
@@ -372,6 +304,9 @@ def run(args):
     # initialize logger
     logger.info('{0} Started'.format(name))
 
+    # create some variables
+
+
     # outfile name
     if args.outfile:
         out_filename = args.outfile
@@ -381,6 +316,7 @@ def run(args):
     # specify output directory
     outdir = args.outdir
 
+
     # create output directory
     make_directory(outdir)
 
@@ -417,68 +353,43 @@ def run(args):
                 fpath = '%s/%s' % (directory, f)
                 _parse_pdf(fpath)
 
+    # simply generate html report from json outfile
+    elif args.gen_html_report:
+        fr = open(args.gen_html_report,'r')
+        analysis_dict = json.loads(fr.read())
+        if create_html_report(analysis_dict, outdir,out_filename):
+            logger.info('Successfully created html report') 
+            sys.exit(0)
+        else:
+            sys.exit(1)
+
     else:
         print('[-] Dunno what to do, bro. Use help. {0} -h'.format(sys.argv[0]))
+        sys.exit(1)
 
-    # move analysis dictionary in queue back to dictionary
-    analysis_dict = {}
-    while ana_q.empty() == False:
-        item = ana_q.get()
-        # print('item ', item)
-        analysis_dict.update(item)
+    # creating the analysis dictionary for reporting
+    analysis_dict = prepare_analysis_dict(ana_q)
 
-    #print('dict:',analysis_dict)
-    # ana_q is empty now
+    # lets go through the different reporting types
+    if args.report_txt:
+        if create_txt_report(analysis_dict, outdir,out_filename):
+            logger.info('Successfully created txt report') 
 
-    # create txt output
-    sep = '-' * 80 + '\n'
-    txtout = "%s/%s.txt" % (outdir, out_filename)
-    fwtxt = open(txtout, 'w')
-    # print(analysis_dict)
-    for k in analysis_dict.keys():
-        fwtxt.write(sep)
-        fname = 'File: %s\n' % (analysis_dict[k]['filename'])
-        ddata = analysis_dict[k]['data']
-        fwtxt.write(fname)
-        for kdata in ddata.keys():
-            metatxt = '%s:%s\n' % (kdata, ddata[kdata])
-            fwtxt.write(metatxt)
-        fwtxt.write(sep)
-    fwtxt.close()
+    if args.report_json:
+        if create_json_report(analysis_dict, outdir,out_filename):
+            logger.info('Successfully created json report') 
 
-    # create json output
-    jsonout = "%s/%s.json" % (outdir, out_filename)
-    fwjson = open(jsonout, 'w')
+    if args.report_html:
+        if create_html_report(analysis_dict, outdir,out_filename):
+            logger.info('Successfully created html report') 
 
-    # print(analysis_dict)
-    jdata = json.dumps(analysis_dict)
-    fwjson.write(jdata)
-    fwjson.close()
+    if args.report_url_txt:
+        if create_url_txt(url_d, outdir,out_filename):
+            logger.info('Successfully created txt url report') 
 
-    # create html from json
-    htmlout = "%s/%s.html" % (outdir, out_filename)
-    fwhtml = open(htmlout,'w')
-    #print(jdata)
-    html = json2html.convert(json = jdata)
-    fwhtml.write(html)
-    fwhtml.close()
-    
-
-    # create url savefile
-    # print('url_d: ', url_d)
-    jsonurlout = "%s/%s_url.json" % (outdir, out_filename)
-    fwjson = open(jsonurlout, 'w')
-    jdata = json.dumps(url_d)
-    fwjson.write(jdata)
-    fwjson.close()
-
-    txtout = "%s/%s_url.txt" % (outdir, out_filename)
-    fwtxt = open(txtout, 'w')
-    for k in url_d.keys():
-        ddata = url_d[k]
-        metatxt = '%s:%s\n' % (ddata['url'], ddata['filename'])
-        fwtxt.write(metatxt)
-    fwtxt.close()
+    if args.report_url_json:
+        if create_url_json(url_d, outdir,out_filename):
+            logger.info('Successfully created json url report') 
 
     return 42
 
@@ -504,8 +415,14 @@ def main():
                         help="specify domain or tld to scrape for pdf-files", default=None)
     parser.add_argument('-sn', '--search-number', action='store', dest='search_stop', required=False,
                         help="specify how many files are searched", default=10, type=int)
-    parser.add_argument('-z', '--disable-cert-check', action='store_false', dest='cert_check', required=False,
-                        help="if the target domain(s) run with old or bad certificates", default=True)
+    parser.add_argument('-z', '--disable-cert-check', action='store_false', dest='cert_check', required=False,help="if the target domain(s) run with old or bad certificates", default=True)
+                        
+    parser.add_argument('-ghr', '--gen-html-report', action='store', dest='gen_html_report', required=False,help="If you want to generate the html report after editing the json outfile (parameter: pdfgrab_analysis.json)")
+    parser.add_argument('-rtd', '--report-text-disable', action='store_false', dest='report_txt', required=False,help="Disable txt report",default=True)
+    parser.add_argument('-rjd', '--report-json-disable', action='store_false', dest='report_json', required=False,help="Disable json report",default=True)
+    parser.add_argument('-rhd', '--report-html-disable', action='store_false', dest='report_html', required=False,help="Disable html report",default=True)
+    parser.add_argument('-rutd', '--report-url-text-disable', action='store_false', dest='report_url_txt', required=False,help="Disable url txt report",default=True)
+    parser.add_argument('-rujd', '--report-url-json-disable', action='store_false', dest='report_url_json', required=False,help="Disable url json report",default=True)
 
     if len(sys.argv)<2:
         parser.print_help(sys.stderr)