Merge branch 'master' of http://192.168.1.6/phowell/canvasapp

apphelp.py

@@ -71,22 +71,21 @@ prog_write.close()
 
 ## Functions
 #
-# Total TODOs remaining: 57
+# Total TODOs remaining: 67
 #
 # TODOs per file:  
 #
-#	1 - checker.py 
+#	6 - content.py 
-#	1 - content.py 
 #	6 - courses.py 
 #	3 - curriculum.py 
-#	5 - depricated.py 
+#	6 - depricated.py 
-#	6 - localcache.py 
+#	7 - localcache.py 
 #	2 - outcomes.py 
-#	20 - pipelines.py 
+#	17 - pipelines.py 
 #	2 - server.py 
-#	2 - tasks.py 
+#	5 - tasks.py 
 #	1 - tempget.py 
-#	8 - users.py 
+#	12 - users.py 
 
 #
 
@@ -95,8 +94,9 @@ prog_write.close()
 
 __init__.py
 
+canvas_secrets.py
+
 checker.py
-		todo: make this sweet
 
 	def safe_html(html):
 
@@ -120,12 +120,6 @@ content.py
 
 	def d(s):
 
-	def stripper(s):
-
-	def mycleaner(s):
-
-	def freshdesk():
-
 	# Build a master file with the entire class content
 	def accessible_check(id=""):
 		todo: include linked pages even if they aren't in module
@@ -140,12 +134,6 @@ content.py
 	# DL pages only
 	def grab_course_pages(course_num=-1):
 
-	# Appears to not be used
-	def put_course_pages():
-
-	# Also not used
-	def put_revised_pages():
-
 	# Download, clean html, and reupload page
 	def update_page():
 
@@ -163,13 +151,32 @@ content.py
 
 	def multiple_downloads():
 
+	def demo_vector_search():
+
+	    def is_complete_sentence(text):
+		todo: site scraper
+		todo: find package that extracts text from web page
+		todo: master list of what to index.
+		todo: PDFs and DOCXs
+		todo: fix urls w/ anchors
+
+	def clean_fn(s):
+
+	def format_html(html):
+
+	    def visit(self, link, source=None):
+
+	    def fail(self, link):
+
+	def crawl():
+
+	def txt_clean_index():
+
+	def samples():
+
 courses.py
 		todo: 
 
-	def int_or_zero(x):
-
-	def float_or_zero(x):
-
 	# Gott 1 Bootcamp - report on who completed it.
 	def get_gott1_passers():
 
@@ -179,15 +186,12 @@ courses.py
 	# Who, in a class, passed?
 	def get_course_passers(course, min_passing, passers_filename, still_active_filename):
 
+	# Who, in a class and a quiz, passed?
+	def get_quiz_passers():
+
 	# Change courses to show 2 announcements
 	def change_course_ann_homepage(id="10458"):
 
-	def scrape_bookstore():
-		todo: where does the most recent schedule come from?
-
-	# Input: xxxx_sched.json. Output: xxxx_latestarts.txt    
-	def list_latestarts():
-
 	# All students enrolled in a class in the given semester. Simpler verson of below. Return SET of course_ids.
 	def users_in_semester():
 		todo: 
@@ -203,13 +207,18 @@ courses.py
 	def getTerms(printme=1, ask=1):
 		todo: unsafe overwrite
 
-	def getCourses():   # a dict
+	def getCourses(x=0):   # a dict
+
+	def update_course_conclude(courseid="13590",enddate='2021-12-23T01:00Z'):
 
 	# Relevant stuff trying to see if its even being used or not
-	def course_term_summary():
+	def course_term_summary_local(term="176",term_label="FA22"):
+
+	# Relevant stuff trying to see if its even being used or not
+	def course_term_summary(term="176",term_label="FA22"):
 
 	# Fetch all courses in a given term
-	def getCoursesInTerm(term=0,show=1,active=0):        # a list
+	def getCoursesInTerm(term=0,get_fresh=1,show=0,active=0):        # a list
 
 	def getCoursesTermSearch(term=0,search='',v=0):
 
@@ -217,6 +226,16 @@ courses.py
 
 	def xlistLineSummary(c,sections={}):
 
+	def numbers_in_common(L):
+
+	def combined_name(nic,L):
+
+	def semester_cross_lister():
+
+	def xlist_ii(parasite_id,host_id,new_name,new_code):
+
+	def all_semester_course_sanity_check():
+
 	def eslCrosslister():
 
 	def xlist(parasite='', host=''):        # section id     ,    new course id 
@@ -226,21 +245,93 @@ courses.py
 
 	def enroll_stem_students_live():
 
+	def enroll_bulk_students_bydept(course_id, depts, the_term="172", cautious=1):     #  a string,  a list of strings
+		todo: not done here
+
+	def enroll_art_students_live():
+
 	def enroll_orientation_students():
 
-	def summarize_proportion_online_classes(u):
+	def enroll_o_s_students():
-
-	def summarize_num_term_classes(u):
 
 	def make_ztc_list(sem='sp20'):
 
 	def course_search_by_sis():
 
+	def mod_eval_visibility( shell_id, visible=True ):
+
+	def instructor_list_to_activate_evals():
+
 	def add_evals(section=0):
+
+	def course_dates_terms(section=0):
+
+	def remove_n_analytics(section=0):
+
+	def create_sandboxes():
+
+	def course_term_summary_2():
+
+	def get_ext_tools():
+
+	def set_ext_tools():
 		todo: wanted: group shell for each GP (guided pathway) as a basic student services gateway....
 
+cq_demo.py
+
+	def fetch(target):
+
+curric2022.py
+
+	def fetch_all_programs():
+
+	def nothing(x=0):
+
+	def clean(st):
+
+	def recur_matcher(item, depth=0):
+
+	def single_course_parse(c):
+
+	def match_style_test():
+
+	def single_program_path_parse(c):
+
+	def path_style_prog():
+
+	def term_txt_to_code(t):
+
+	def all_outcomes():
+
+	def ddl():
+
+	def splitclassline(cl, id=''):
+
+	def path_style_2_html():
+
+	def course_path_style_2_html():
+
+	def another_request(url,startat):
+
+	def fetch_all_classes():
+
+	def recur_path_matcher(item, path=[]):
+
+	    def x2_path_update(x,y,z):
+
+	def pathstyle(theclass):
+
+	def single_course_path_parse(c):
+
+	def path_style_test():
+
+	def make_sl():
+
+	def course_rank():
+
 curriculum.py
-		todo: These secrets
+
+	def dbg(x): 
 
 	def another_request(url,startat):
 
@@ -335,6 +426,9 @@ curriculum.py
 
 	        def is_online_inblock(c):
 
+	# 9/2021 clean programs to good json
+	def organize_programs_stage2():
+
 	# of all the programs, what can be accomplished online?    
 	def find_online_programs():
 
@@ -408,6 +502,35 @@ curriculum.py
 
 	def cq_8020_start():
 
+	def recurse3(sec,path=''):
+
+	def get_id_sortorder(sec):
+
+	def include_exclude(str,inc,exc=[]):
+
+	def pbd3(str):
+
+	def handleField(f):
+
+	def boolToStr(b):
+
+	# Almost final formatting
+	def prog_info_to_entry(c):
+
+	def cbd_to_entry(c):
+
+	def pc5(str):
+
+	def remove_prefix(str,i):
+
+	def course_to_entry(c,order="0"):
+
+	def courseline_to_pretty(line):
+
+	# restarted oct 2019 and try to simplify
+	def prog_take_4(program):
+		todo: 
+
 curriculum2020.py
 
 	def to_md(s):
@@ -473,6 +596,18 @@ curriculum_patterns.py
 	def jj2(a,b,c,d):
 
 depricated.py
+		todo: where does the most recent schedule come from?
+
+	# Input: xxxx_sched.json. Output: xxxx_latestarts.txt    
+	def list_latestarts():
+
+	def prep_online_courses_df():
+
+	def course_is_online(crn):
+
+	def get_crn_from_name(name):
+
+	def get_enrlmts_for_user(user,enrollments):
 
 	# Don't know
 	def demo():
@@ -504,10 +639,30 @@ depricated.py
 	    def sd():
 
 	def serve():
+
+	def summarize_proportion_online_classes(u):
+
+	def summarize_num_term_classes(u):
 		todo: this duplicates courses.py  ??
 
 	# Prompt for course id, return list of user dicts.   TODO this duplicates courses.py  ??
 	def getUsersInCourse(id=0):   # returns list
+
+	def recur_look_for_leafs(item,indent=0,show=1):
+
+	def am_i_a_leaf(item):
+
+	def sampleclass():
+
+	def matchstyle():
+
+	def user_role_and_online():
+
+	def more_unused_xreferencing():
+
+	def users_p_file():
+
+	    def com_channel_dim():
 		todo: 
 
 	# NO LONGER USED - SEE COURSES
@@ -559,6 +714,22 @@ depricated.py
 
 	def get_schedule(term='201870', sem='fall'):
 
+	def dates(s):
+
+	def parse_www_csv_sched():
+
+	def parse_json_test_sched():
+
+	def put_revised_pages():
+
+	def put_course_pages():
+
+	def freshdesk():
+
+gpt.py
+
+graphics.py
+
 interactive.py
 
 	def dict_generator(indict, pre=None):
@@ -579,8 +750,24 @@ interactive.py
 
 	    def before_request():
 
+	    def clears():
+
+	    def dpi():
+
+	    def dpi2():
+
+	    def screenoff_a():
+
+	    def light():
+
+	    def do_image(filename):
+
+	    def do_image_crop(filename,x,y,w,h,newname):
+
 	    def save_post():
 
+	    def writing_img(fname):
+
 	    def restart():
 
 	    def dispatch3(func,arg,arrg):
@@ -597,12 +784,12 @@ interactive.py
 
 	    def home():
 
-	    def send_jslib(path):
-
 	    def send_cachedata(path):
 
 	    def send_js(path):
 
+	    def send_jslib(path):
+
 	    def s(key,val):
 
 	    def do_sample():
@@ -787,8 +974,6 @@ interactivex.py
 
 	def repl():
 
-ipython_log.py
-
 localcache.py
 
 	def db():
@@ -893,7 +1078,12 @@ localcache.py
 
 	    def f(x):
 
-	# get student count and teacher name from local db
+	def get_courses_in_term_local(term="172"):
+
+	# get student count
+	def course_student_stats(canvasid):
+
+	# get teacher name from local db
 	def course_quick_stats(canvasid):
 
 	# What a student has taken / teacher has taught
@@ -932,24 +1122,37 @@ localcache.py
 
 	    def qstrip(txt): return txt.strip('"')
 
-	def more_unused_xreferencing():
-
-	def user_role_and_online():
-
 	def comm_channel_file():
 
 	def pseudonym_file():
 
-	def users_p_file():
-
-	    def com_channel_dim():
-
 	def abcd():
 
 	def crns_to_teachers():
 
+	def all_sem_courses_teachers():
+
+	def to_sis_sem(s):
+
+	def build_db_schedule():
+
+	    def finder(st):
+
+	def process_enrollment_data():
+
+	    def sem_to_idx(s): 
+		todo: 
+
+	def do_encoding():
+
 main.py
 
+myconsole.py
+
+	def handler(signum, frame):
+
+	def mainloop():
+
 outcomes.py
 
 	def outcome_overview(term=21):
@@ -958,12 +1161,10 @@ outcomes.py
 
 	def connect_acct_oc_to_course(course_id,oc_group_id):
 
-	def outcome_groups():
+	def outcome_groups_dump():
 
 	def outcome_groups_backup():
 
-	def x_ref_dept_names():
-
 	def create_course_group(short,parent):
 
 	def create_dept_group(short):
@@ -992,6 +1193,74 @@ outcomes.py
 
 	def slo_source_by_dept():
 
+	def printj(j):
+
+	def writej(o,j):    
+
+	# Get root outcome group
+	def root_og():
+
+	def recur_og():
+
+	def recur_main(out,g_url=""):
+
+	def recur2(out,og={}):
+
+	def all_og():
+
+	def course_slo_getter(q):
+
+	def threaded_getter():
+
+	def demo_o_fetch():
+
+	def outcome_groups_2021():
+
+	def x_ref_dept_names():
+
+	def all_outcome_results_in_term(termid=''):
+
+	def all_outcome_results_in_term_sub(termid=''):
+
+	def all_linked_outcomes_in_term(termid=''):
+
+	def all_linked_outcomes_in_term_sub(termid=''):
+
+	def assemblerow(g,parent=''):
+
+	def recur_full_fetch(out,g,parent=""):
+
+	# return the appropriate cq course version.
+	def find_cq_course_version(code):
+
+	def outcome_groups():
+
+	def summary_string(s):
+
+	def add_outcomes_course_id(canvas_id):
+
+	def add_outcomes_course_code():
+
+	def add_outcomes_course_code_sub(target_code='AJ184',term=178,fresh=0):
+
+	def add_csis_sp22():
+
+	def quick_add_course_outcomes(ilearn_course_id, cq_outcome_id_list):
+
+	def stringpad(s,n):
+
+	def code_from_ilearn_name(n,verbose=0):
+
+	def parse_ilearn_course_names_ALLSEMESTERS():
+
+	def parse_ilearn_course_names(term='178',fresh=1,log=0):
+
+outcomes2022.py
+
+	def course_slo_getter(q):
+
+	def ilearn_shell_slo_to_csv(shell_slos):
+
 patterns_8020.py
 
 patterns_topdown.py
@@ -1041,11 +1310,10 @@ patterns_topdown.py
 	def jj2(a,b,c,d):
 
 pipelines.py
-		todo: secrets
 		todo: all these constants for SSB  --  line 1008
-		todo: secrets
+		todo: https://stackoverflow.com/questions/42656247/how-can-i-use-canvas-data-rest-api-using-python
 
-	def d(s): 
+	def d(s,end=''): 
 
 	# Main canvas querying fxn    
 	def fetch(target,verbose=0):
@@ -1073,12 +1341,6 @@ pipelines.py
 	def getSemesterSchedule(short='sp21'):                # I used to be current_schedule
 		todo: Some semesters have a different format.... partofday   type   site  xxx i just dL'd them again
 
-	def prep_online_courses_df():
-
-	def course_is_online(crn):
-
-	def get_crn_from_name(name):
-
 	def get_enrlmts_for_user(user,enrollments):
 
 	# Get something from Canvas Data
@@ -1110,6 +1372,7 @@ pipelines.py
 	### course is a list of 1-3 lists, each one being a line in the schedule's output. First one has section     
 	def course_start(course):
 		todo: use this to make a early/late/short field and store semester dates w/ other constants
+		todo: do these years matter?
 
 	def time_to_partofday(t):
 		todo: account for multiple sites/rows
@@ -1132,9 +1395,10 @@ pipelines.py
 
 	# Use Firefox and log in to ssb and get full schedule. Only works where selenium is installed
 	def scrape_schedule():
-		todo: my data here.... secret
 		todo: 
 
+	def dza_sched():
+
 	# recreate schedule json files with most current online schedule format. 
 	def recent_schedules():
 		todo: sems is a global in this file. Is that the right thing to do?   
@@ -1149,7 +1413,6 @@ pipelines.py
 
 	# From instructure sftp site
 	def fetch_current_rosters():
-		todo: secret
 
 	def fetch_current_rosters_auto():
 
@@ -1159,7 +1422,6 @@ pipelines.py
 
 	# Upload a json file to www
 	def put_file(remotepath,localpath, localfile,prompt=1):
-		todo: remove this secret
 		todo: these paths
 
 	def sec(t): return "<h3>"+t+"</h3>\n"
@@ -1205,8 +1467,43 @@ pipelines.py
 
 	def scrape_schedule_py():
 
+	def scrape_schedule_multi():
+
+	def scrape_for_db():
+
+	def argos_data():
+
+	def days_times(s):
+
+	def remove_year(s):
+
+	def argos_data_from_cvc():
+
+	def expand_old_semesters():
+
+	# Input: xxxx_sched.json. Output: xxxx_latestarts.txt    
+	def list_latestarts(term="su23"):
+
 server.py
 
+	def mqtt_loop():
+
+	# called when MQTT server connects
+	def on_connect(client, userdata, flags, rc):
+
+	# The callback for when a PUBLISH message is received from the server.
+	def on_message(client, userdata, msg):
+
+	def displaypi_on():
+
+	def displaypi_off():
+
+	def desklight():
+
+	def clearscreens():
+
+	def screenoff():
+
 	def tag(x,y): return "<%s>%s</%s>" % (x,y,x)
 
 	def tagc(x,c,y): return '<%s class="%s">%s</%s>' % (x,c,y,x)
@@ -1242,6 +1539,8 @@ server.py
 
 	def randPic():
 
+	def do_img_crop(im):
+
 	def sample():
 
 	def sample2(a=""):
@@ -1286,6 +1585,16 @@ server.py
 
 	def staff_dir(search=''):
 
+	def find_goo(n):
+
+	def byname(x): 
+
+	def fn_to_struct( n, staff ):
+
+	def image_edit(filename=''):
+
+	def image_crop(filename,x,y,w,h,newname=''):
+
 	def server_save(key,value):
 
 	def server_dispatch_json(function_name,arg='', arg2=''):
@@ -1310,6 +1619,8 @@ stats.py
 
 tasks.py
 
+	def scrape_bookstore():
+
 	def survey_answer(q=0):
 
 	def survey_organize():
@@ -1360,6 +1671,36 @@ tasks.py
 
 	def pos_atten():
 
+	def lname(x):
+
+	def l_initial(x):
+
+	def job_titles2():
+
+	def job_titles():
+
+	# an early version, before tearing up...
+	def job_titles3():
+
+	def index_pics():
+
+	def cmtes():
+
+	def strip(x): return x.strip()
+
+	def esc_comma(x): return re.sub(',','[CMA]',x)
+
+	def by_sem(x): return x['sem']
+
+	def parse_schedule():
+		todo: check if i need to update it
+		todo: some weird hour offset issue w/ these activities
+
+	def cal():
+		todo: >
+
+	def file_renamer():
+
 temp.py
 
 tempget.py
@@ -1466,6 +1807,7 @@ users.py
 	def teacher_basic_info(sched, from_ilearn, names):
 
 	    def find_that_name(x):
+		todo: Old and broken
 
 	# Outputs:   cache/teacher_by_semester.csv, 
 	def teacherModalityHistory(sched=[],names=[]):
@@ -1475,6 +1817,7 @@ users.py
 
 	# Outputs:   cache/course_teacher_combos.csv, 
 	def teacherSharedCourses(a=[]):
+		todo: this is broken
 
 	# How many courses in each department were taught in the last year?
 	def departmentCountCourses(a=[]):
@@ -1512,6 +1855,7 @@ users.py
 
 	# Make one big csv file of everything I know about a teacher    
 	def getTeachersInfoMain():
+		todo: - broken
 
 	def enroll_staff_shell():
 
@@ -1521,8 +1865,8 @@ users.py
 
 	# Get views counts on current teachers. todo: month is hardcoded here
 	def get_recent_views(id=1):
+		todo: broken?
 
-	# Have they taught online or hybrid classes?        
 	def categorize_user(u):
 		todo: threaded
 
@@ -1539,8 +1883,6 @@ users.py
 	# Go through my local profile pics, upload any that are missing.
 	def uploadPhoto():
 
-	def test_email():
-
 	def create_ztc_list():
 
 	def get_user_info(id):
@@ -1583,8 +1925,20 @@ users.py
 
 	def one_course_enrol():
 
+	def find_new_teachers():
+
+	def user_db_sync():
+
+	def find_no_goo():
+
+	def track_a_user():
+
 util.py
 
+	def stripper(s):
+
+	def mycleaner(s):
+
 	def print_table(table):
 
 	def remove_nl(str):
@@ -1597,6 +1951,10 @@ util.py
 
 	def clean_title(st):    
 
+	def int_or_zero(x):
+
+	def float_or_zero(x):
+
 	def match59(x):
 
 	def item_2(x): return x[2]

content.py

@@ -1,17 +1,18 @@
 
 
 #saved_titles = json.loads( codecs.open('cache/saved_youtube_titles.json','r','utf-8').read() )
-import requests, codecs, os, re, json
+import requests, codecs, os, re, json, sys, pypandoc
+import webbrowser, bs4, trafilatura, pickle, tomd, checker 
+import html2markdown as h2m
 from pipelines import header, fetch, url, put_file
 from util import clean_title, to_file_friendly, minimal_string, stripper, mycleaner
 from bs4 import BeautifulSoup as bs
 from html.parser import HTMLParser
-import tomd, checker 
+from collections import defaultdict
-import html2markdown as h2m
+from pdfminer.high_level import extract_text
-import pypandoc
+from sentence_transformers import SentenceTransformer, util
-import webbrowser
-h = HTMLParser()
 
+h = HTMLParser()
 
 DBG = 1
 
@@ -21,8 +22,6 @@ def d(s):
 
 
 
-
-
 # Download everything interesting in a course to a local folder
 # Build a master file with the entire class content
 def accessible_check(id=""):
@@ -828,64 +827,266 @@ Schedule an In-Person, Phone or Zoom Appointment"""
     print(f"Vector for the word '{example_word}': {vector}")
 
 
+
+def makedir():
+    files = os.listdir('cache/crawl')
+    #print(files)
+    files.sort()
+    for f in files:
+        m = re.match(r'https?..www\.gavilan\.edu\+(.*)\.\w\w\w\w?\.txt$',f)
+        if m:
+            name = m.groups()[0]
+            parts = name.split('+')
+            print(parts)
+
+def manual_index():
+    files = os.listdir('cache/crawl')
+    #print(files)
+    ii = codecs.open('cache/crawl/index.html','w','utf-8')
+    ii.write('<html><body><h1>Site index</h1>\n')
+    files.sort()
+    for f in files:
+        m = re.match(r'https?..www\.gavilan\.edu\+(.*)\.\w\w\w\w?\.txt$',f)
+        if m:
+            name = m.groups()[0]
+            parts = name.split('+')
+            ii.write('<br /><a href="mirror/'+f+'">'+f+'</a>\n')
+
+def my_site():
+    files = os.listdir('cache/crawl')
+    output = []
+    files.sort()
+    for f in files:
+        m = re.match(r'https?..www\.gavilan\.edu\+(.*)\.\w\w\w\w?\.txt$',f)
+        if m:
+            name = m.groups()[0]
+            parts = name.split('+')
+            output.append(parts)
+    return output
+
+
 ## TODO  site scraper
-
+## TODO  find package that extracts text from web page
-## TODO  finde package that extracts text from web page
-
 ### TODO master list of what to index.
 
-from pattern.web import URL, plaintext, extension
+## TODO PDFs and DOCXs
-from pattern.web import download
+## TODO fix urls w/ anchors
-from pattern import URL, MIMETYPE_IMAGE
-from pattern.web import Crawler
-from util import clean_title
-
-save_folder = 'cache/crawl'
-
-class GavCrawl(Crawler): 
-    def visit(self, link, source=None):
-        print 'visited:', repr(link.url), 'from:', link.referrer
-        txt = plaintext(link.source)  ## , keep={'h1':[], 'h2':[], 'strong':[], 'a':['href']})
-        codecs.open(save_folder + '/' + clean_title(link.url) + '.txt').write(txt)
-
-    def fail(self, link):
-        print 'failed:', repr(link.url)
 
 def crawl():
-    p = GavCrawl(links=['http://www.gavilan.edu/'], delay=3)
+    import scrapy, logging
-    while not p.done:
+    from scrapy.crawler import CrawlerProcess
-        p.crawl(method=DEPTH, cached=False, throttle=3)
+
+    logger = logging.getLogger()
+    logger.setLevel(level=logging.CRITICAL)
+    logging.basicConfig(level=logging.CRITICAL)
+    logger.disabled = True
+
+
+    avoid = ['ezproxy','community\.gavilan\.edu','archive\/tag','archive\/category', 'my\.gavilan\.edu',  'augusoft', 
+             'eis-prod', 'ilearn\.gavilan', 'mailto', 'cgi-bin', 'edu\/old\/schedule',  ]
+
+    class MySpider(scrapy.Spider):
+        name = 'myspider'
+        #start_urls = ['https://gavilan.curriqunet.com/catalog/iq/1826']
+        start_urls = ['https://www.gavilan.edu']
+
+
+        """
+        logging.getLogger("scrapy").setLevel(logging.CRITICAL)
+        logging.getLogger("scrapy.utils.log").setLevel(logging.CRITICAL)
+        logging.getLogger("scrapy.extensions.telnet").setLevel(logging.CRITICAL)
+        logging.getLogger("scrapy.middleware").setLevel(logging.CRITICAL)
+        logging.getLogger("scrapy.core.engine").setLevel(logging.CRITICAL)
+        logging.getLogger("scrapy.middleware").setLevel(logging.CRITICAL)
+
+        logger.disabled = True"""
+
+        def parse(self, response):
+            print('visited:', repr(response.url), 'status:', response.status)
+
+            if re.search(r'\.pdf$', response.url):
+                m = re.search(r'\/([^\/]+\.'+ext+')$', response.url)
+                if m:
+                    print("saving to ", save_folder + '/' + clean_fn(response.url))
+                    pdf_response = requests.get(response.url)
+                    with open(save_folder + '/' + clean_fn(response.url), 'wb') as f:
+                        f.write(pdf_response.content)
+                    text = extract_text(save_folder + '/' + clean_fn(response.url))
+                    codecs.open(save_folder + '/' + clean_fn(response.url) + '.txt','w','utf-8').write(text)
+
+            for ext in ['doc','docx','ppt','pptx']:
+                if re.search(r'\.'+ext+'$', response.url):
+                    m = re.search(r'\/([^\/]+\.'+ext+')$', response.url)
+                    if m:
+                        print("saving to ", save_folder + '/' + clean_fn(response.url))
+                        pdf_response = requests.get(response.url)
+                        with open(save_folder + '/' + clean_fn(response.url), 'wb') as f:
+                            f.write(pdf_response.content)
+                        #text = extract_text(save_folder + '/' + clean_fn(response.url) + '.txt')
+                        output = pypandoc.convert_file(save_folder + '/' + clean_fn(response.url), 'html', extra_args=['--extract-media=%s' % hash ]) 
+                        txt_output = trafilatura.extract(response.text,include_links=True, deduplicate=True, include_images=True, include_formatting=True)
+                        if txt_output:
+                            codecs.open(save_folder + '/' + clean_fn(response.url) + '.txt','w','utf-8').write(txt_output)
+
+            for ext in ['jpg','jpeg','gif','webp']:
+                if re.search(r'\.'+ext+'$', response.url):
+                    m = re.search(r'\/([^\/]+\.'+ext+')$', response.url)
+                    if m:
+                        print("saving to ", save_folder + '/' + clean_fn(response.url))
+                        pdf_response = requests.get(response.url)
+                        with open(save_folder + '/' + clean_fn(response.url), 'wb') as f:
+                            f.write(pdf_response.content)
+
+            f_out = codecs.open(save_folder + '/' + clean_fn(response.url) + '.txt','w','utf-8')
+
+            this_output = trafilatura.extract(response.text,include_links=True, deduplicate=True, include_images=True, include_formatting=True)
+            if this_output:
+                f_out.write(this_output)
+            f_out.close()
+            links = response.css('a::attr(href)').getall()
+
+            # Follow each link and parse its contents
+            
+            for link in links:
+                go = 1
+                full_link = response.urljoin(link)
+                print('++++++ trying ', full_link)
+
+                if not re.search(r'gavilan\.edu',full_link):
+                    go = 0
+                    print('--- not gav edu')
+                else:
+                    if re.search(r'hhh\.gavilan\.edu',full_link):
+                        pass
+                    elif not re.search(r'^https?:\/\/www\.gavilan\.edu',full_link):
+                        # need to add www to gavilan.edu
+                        m = re.search(r'^(https?:\/\/)gavilan\.edu(\/.*)$',full_link)
+                        if m:
+                            full_link = m.group(1) + 'www.' + m.group(2)
+                for a in avoid:
+                    if re.search(a,full_link):
+                        go = 0
+                        print('--- avoid ', a)
+
+                if go: yield scrapy.Request(full_link, callback=self.parse,
+                                     headers={"User-Agent": "Mozilla/5.0 (iPad; CPU OS 12_2 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Mobile/15E148"})
+                else:
+                    print("------ avoiding ", full_link)
+    # Instantiate a CrawlerProcess object
+    process = CrawlerProcess()
+
+    # Add the MySpider spider to the process
+    process.crawl(MySpider)
+
+    # Start the process
+    logging.basicConfig(level=logging.CRITICAL)
+    logging.getLogger('scrapy').propagate = False
+    logging.getLogger("trafilatura").setLevel(logging.CRITICAL)
+    logging.getLogger("trafilatura").propagate = False
+    logging.getLogger("pdfminer").setLevel(logging.CRITICAL)
+    logging.getLogger("pdfminer").propagate = False
+    logging.getLogger("urllib3").setLevel(logging.CRITICAL)
+    logging.getLogger("urllib3").propagate = False
+    logging.basicConfig(level=logging.CRITICAL)
+    process.start()
 
 
 
-def samples():
+save_folder = 'cache/crawl'
-    crawler = Crawler(links=[], domains=[], delay=20.0, sort=FIFO)
+clean_folder = 'cache/cleancrawl'
 
-    url = URL('http://www.clips.ua.ac.bemedia/pattern_schema.gif')
+def clean_fn(s):
-    print url.mimetype in MIMETYPE_IMAGE
+    s = re.sub(r'[\s:]+','',s)
+    s = re.sub(r'\/','+',s)
+    return s
+    
+def format_html(html):
+    soup = bs4.BeautifulSoup(html, 'html.parser')
+    return soup.prettify()
     
 
-    #html = download('http://www.clips.ua.ac.be/', unicode=True)
-    s = URL('http://www.clips.ua.ac.be').download()
-    s = plaintext(s, keep={'h1':[], 'h2':[], 'strong':[], 'a':['href']})
 
 
-    # getting absolute urls
+def txt_clean_index():
-    from pattern.web import URL, DOM, abs
+    files = os.listdir(save_folder)
+    line_freq = defaultdict(int)
     
-    url = URL('http://www.clips.ua.ac.be')
+    # first pass
-    dom = DOM(url.download())
+    for f in files:
-    for link in dom('a'):
+        lines = codecs.open(save_folder + '/' + f,'r','utf-8').readlines()
-        print abs(link.attributes.get('href',''), base=url.redirect or url.string) 
+        for L in lines:
+            L = L.strip()
+            line_freq[L] += 1
     
-    # get pdfs
+    # second pass
-    from pattern.web import URL, PDF
+    for f in files:
+        print("\n\n",f)
+        lines = codecs.open(save_folder + '/' + f,'r','utf-8').readlines()
+        out = codecs.open(clean_folder + '/' + f,'w','utf-8')
+        for L in lines:
+            L = L.strip()
+            if L in line_freq and line_freq[L] > 3:
+                continue
+            print(L)
+            out.write(L + '\n')
+        out.close()
 
-    url = URL('http://www.clips.ua.ac.be/sites/default/files/ctrs-002_0.pdf')
-    pdf = PDF(url.download())
-    print pdf.string
 
 
+def search_embeddings():
+    model = SentenceTransformer('all-MiniLM-L6-v2')
+    save_embeds = pickle.load( open( "cache/embeddings.p", "rb" ) )
+    columns = list(zip(*save_embeds))
+    files = columns[0]
+    sentences = columns[1]
+    embeddings = columns[2]
+
+    print(files[:20])
+    print(sentences[:20])
+    print(embeddings[:20])
+
+    s = ''
+    while s != 'q':
+        s = input("search or 'q' to quit: ")
+        if s == 'q':
+            return
+        query_embedding = model.encode(s)
+
+        # Compute the cosine similarity between the query embedding and the sentence embeddings
+        cosine_scores = util.cos_sim(query_embedding, embeddings)
+
+        # Sort the sentences by their cosine similarity to the query sentence
+        results = sorted(zip(sentences, cosine_scores, files), key=lambda x: x[1], reverse=True)
+
+        # Print the top 5 results
+        for i, (sentence, score, file) in enumerate(results[:5]):
+            print(f'Top {i+1}: {file} - {sentence} - (Score: {score})')
+
+
+def create_embeddings():
+    model = SentenceTransformer('all-MiniLM-L6-v2')
+    files = os.listdir('cache/crawl')
+    output = []
+    save_embeds = []    #  ['file','sentence','embedding']
+    files.sort()
+    for f in files:
+        m = re.match(r'https?..www\.gavilan\.edu\+(.*)\.\w\w\w\w?\.txt$',f)
+        if m:
+            lines = codecs.open('cache/crawl/' + f,'r','utf-8').readlines()
+            lines = [L.strip() for L in lines]
+            lines = [L for L in lines if L]
+            embeddings = model.encode(lines)
+
+            print("\n-----", f)
+
+            #Print the embeddings
+            for sentence, embedding in zip(lines, embeddings):
+                print("Sentence:", sentence)
+                #print("Embedding:", embedding)
+                
+                save_embeds.append([f,sentence,embedding])
+    pickle.dump( save_embeds, open( "cache/embeddings.p", "wb" ) )
+
 
 if __name__ == "__main__":
     
@@ -897,8 +1098,19 @@ if __name__ == "__main__":
                 # 5: ['import freshdesk content', freshdesk ],
                 6: ['download all a courses pages', grab_course_pages],
                 7: ['demo vector search', demo_vector_search],
+                8: ['crawl',crawl],
+                9: ['clean text index', txt_clean_index],
+               10: ['make web dir struct', manual_index],
+               11: ['create search embeddings', create_embeddings],
+               12: ['do a search', search_embeddings],
               }
     
+    if len(sys.argv) > 1 and re.search(r'^\d+',sys.argv[1]):
+        resp = int(sys.argv[1])
+        print("\n\nPerforming: %s\n\n" % options[resp][0])
+    
+    else:
+        print ('')
         for key in options:
             print(str(key) + '.\t' + options[key][0])
         
@@ -908,5 +1120,3 @@ if __name__ == "__main__":
     # Call the function in the options dict
     options[ int(resp)][1]() 
 
-    
-    

courses.py

@@ -1086,6 +1086,7 @@ def add_evals(section=0):
     s = [ x.strip() for x in codecs.open('cache/sp23_eval_sections.csv','r').readlines()]
     s = list(funcy.flatten(s))
     s.sort()
+    print(s)
     xyz = input('hit return to continue')
     
     #c = getCoursesInTerm(168,0,1)
@@ -1306,7 +1307,6 @@ def set_ext_tools():
     
 if __name__ == "__main__":
     options = { 1: ['Cross check schedule with ztc responses',make_ztc_list] ,  
-                30: ['List latestart classes', list_latestarts ],
                 2: ['Add announcements to homepage', change_course_ann_homepage],
                 3: ['Cross-list classes', xlist ],
                 4: ['List students who passed quiz X', get_quiz_passers],
@@ -1335,6 +1335,7 @@ if __name__ == "__main__":
                 27: ['Fine tune term dates and winter session', course_dates_terms],
                 28: ['Cross list a semester from file', semester_cross_lister],
                 29: ['Check all courses & their sections in semester', all_semester_course_sanity_check],
+                #30: ['List latestart classes', list_latestarts ],
                 # TODO wanted: group shell for each GP (guided pathway) as a basic student services gateway....
                 #
               }

depricated.py

@@ -1807,3 +1807,95 @@ def freshdesk():
 
 
 
+
+
+#### content.py
+
+
+from pattern.web import  plaintext, extension
+from pattern.web import download
+#from pattern import URL, MIMETYPE_IMAGE
+from pattern.web import Crawler, DEPTH, FIFO, MIMETYPE_IMAGE, MIMETYPE_PDF
+
+class GavCrawl(Crawler): 
+    def visit(self, link, source=None):
+        print('visited:', repr(link.url), 'from:', link.referrer)
+        print('   ', link.url.mimetype)
+        #txt = plaintext(source, keep={'h1':[], 'h2':[], 'h3':[], 'h4':[], 'td':[], 'strong':[], 'b':[], 'a':['href'], 'img':['src'], 'ul':[], 'ol':[], 'li':[], 'dd':[], 'dt':[], 'i':[]})
+        #codecs.open(save_folder + '/' + mycleaner(clean_title(link.url)) + '.txt','w','utf-8').write(tomd.convert(txt))
+        
+        codecs.open(save_folder + '/' + clean_fn(link.url) + '.txt','w','utf-8').write(trafilatura.extract(source,include_links=True, deduplicate=True, include_images=True, include_formatting=True))
+        
+
+    def fail(self, link):
+        print('failed:', repr(link.url))
+        if re.search(r'\.pdf$', link.url):
+            m = re.search(r'\/([^\/]+\.pdf)$', link.url)
+            if m:
+                save_file = m.group(1)
+                print("saving to ", save_folder + '/' + save_file)
+                pdf_response = requests.get(link.url)
+                with open(save_folder + '/' + save_file, 'wb') as f:
+                    f.write(pdf_response.content)
+                text = extract_text(save_folder + '/' + save_file)
+                #print(text)
+                codecs.open(save_folder + '/' + save_file + '.txt','w','utf-8').write(text)
+            else:
+                print("no match for pdf url: ", link.url)
+
+        for ext in ['jpg','jpeg','gif','webp']:
+            if re.search(r'\.'+ext+'$', link.url):
+                m = re.search(r'\/([^\/]+\.'+ext+')$', link.url)
+                if m:
+                    save_file = m.group(1)
+                    print("saving to ", save_folder + '/' + save_file)
+                    pdf_response = requests.get(link.url)
+                    with open(save_folder + '/' + save_file, 'wb') as f:
+                        f.write(pdf_response.content)
+                else:
+                    print('no match for '+ext+' url: ', link.url)
+
+def crawl2():
+    #p = GavCrawl(links=['http://www.gavilan.edu/'], domains=['gavilan.edu', 'gavilan.curriqunet.com','www.boarddocs.com'], delay=0.75)
+    #p = GavCrawl(links=['https://gavilan.edu/finaid/2022-23DirectLoanApplication1.pdf'], domains=['gavilan.edu', 'gavilan.curriqunet.com','www.boarddocs.com'], delay=0.75)
+    p = GavCrawl(links=['https://gavilan.curriqunet.com/catalog/iq/1826'], domains=['gavilan.edu', 'gavilan.curriqunet.com','www.boarddocs.com'], delay=0.75)
+
+    
+
+
+    while not p.done:
+        try:
+            p.crawl(method=DEPTH, cached=False, throttle=0.76)
+        except Exception as e:
+            print("Exception: ", e)
+
+
+
+def samples():
+    crawler = Crawler(links=[], domains=[], delay=20.0, sort=FIFO)
+
+    url = URL('http://www.clips.ua.ac.bemedia/pattern_schema.gif')
+    print(url.mimetype in MIMETYPE_IMAGE)
+
+
+    #html = download('http://www.clips.ua.ac.be/', unicode=True)
+    s = URL('http://www.clips.ua.ac.be').download()
+    s = plaintext(s, keep={'h1':[], 'h2':[], 'strong':[], 'a':['href']})
+
+
+    # getting absolute urls
+    from pattern.web import URL, DOM, abs
+
+    url = URL('http://www.clips.ua.ac.be')
+    dom = DOM(url.download())
+    for link in dom('a'):
+        print(abs(link.attributes.get('href',''), base=url.redirect or url.string))
+
+    # get pdfs
+    from pattern.web import URL, PDF
+
+    url = URL('http://www.clips.ua.ac.be/sites/default/files/ctrs-002_0.pdf')
+    pdf = PDF(url.download())
+    print(pdf.string)
+
+

gpt.py

@@ -4,8 +4,8 @@ import openai
 from canvas_secrets import openai_org, openai_api_key
 
 
-openai.organization = "org-66WLoZQEtBrO42Z9S8rfd10M"
+openai.organization = openai_org
-openai.api_key = "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"
+openai.api_key = openai_api_key
 #print(openai.Model.list())
 
 my_prompt = "Write a series of texts trying to sell a pen to a stranger."

interactive.py

@@ -1,4 +1,3 @@
-import curses
 import heapq, re, csv, os, shutil, datetime, urllib
 import itertools, time, markdown, csv, json, os.path, webbrowser, threading
 from functools import wraps
@@ -15,6 +14,20 @@ import localcache
 from server import *
 from canvas_secrets import flask_secretkey
 
+from content import my_site
+
+import socket
+this_host = socket.gethostname()
+
+print('\n\n' + this_host, '\n\n')
+
+has_curses = 0
+if this_host != 'ROGDESKTOP':
+    import curses
+    has_curses = 1
+else:
+    print("Skipping curses stuff")
+
 q = Queue() 
 
 
@@ -25,7 +38,6 @@ PORT_NUMBER = 8080 # Maybe set this to 9000.
 
 datafile = 'lambda.csv'
 
-#writing_path = 'c:/users/peter/Nextcloud/Documents/writing/'
 
 
 ####
@@ -95,6 +107,15 @@ def flask_thread(q):
     
     
 
+    @app.route('/mirror')
+    def mirror():
+        return codecs.open('cache/crawl/index.html','r','utf-8').read()
+    
+    
+    @app.route('/mirror/<filename>')
+    def mirror_file(filename):
+        return markdown.markdown( codecs.open('cache/crawl/'+filename,'r','utf-8').read() ) + \
+            "<pre>" + codecs.open('cache/crawl/'+filename,'r','utf-8').read() + "</pre>"
 
     @app.route('/clearscreens')
     def clears():
@@ -166,6 +187,7 @@ def flask_thread(q):
 
     @app.route('/x/writing/images/<fname>')
     def writing_img(fname):
+        # TODO
         img_path = "/media/hd2/peter_home/Documents/writing_img/"
         print(img_path + fname + " - writing images folder")
         img_ext = fname.split('.')[-1]

server.py

@@ -1,5 +1,5 @@
 import json, codecs, re, markdown, os, pypandoc, striprtf, sqlite3, random, urllib
-import subprocess, html
+import subprocess, html, time
 from striprtf.striprtf import rtf_to_text
 from flask import render_template, Response
 from flask import send_from_directory
@@ -16,8 +16,33 @@ from localcache import arrange_data_for_web, depts_with_classcounts, dept_with_s
 from yattag import Doc
 
 
+import socket
+this_host = socket.gethostname()
+print('\n\n server host: ' + this_host, '\n\n')
+
 LECPATH = "/media/hd2/peter_home_offload/lecture/"
 host = 'http://192.168.1.6:5000'
+news_path = '/media/hd2/peter_home/Documents/scripts/browser/'
+writing_path = '/media/hd2/peter_home/Documents/writing/'
+img_path = '/media/hd2/peter_home/Documents/writing_img/'
+pics_path = '/media/hd2/peter_home/misc/'
+
+
+if this_host == 'ROGDESKTOP':
+    LECPATH = "d:/peter_home_offload/lecture/"
+    host = 'http://192.168.1.7:5000'
+    news_path = 'd:/peter_home/Documents/scripts/browser/'
+    writing_path = 'd:/peter_home/Documents/writing/'
+    img_path = 'd:/peter_home/Documents/writing_img/'
+    pics_path = 'd:/peter_home/misc/'
+
+
+
+
+
+   
+
+
 
 
 import paho.mqtt.client as mqtt
@@ -55,7 +80,7 @@ def on_message(client, userdata, msg):
     print("    %s  mqtt msg: %s    data: %s"  % (now, msg.topic, msg.payload.decode()))
 
 
-
+if 0:
     while(mqtt_offline):
         try:
             client = mqtt.Client()
@@ -114,18 +139,6 @@ def screenoff():
 ######
 
 
-news_path = '/media/hd2/peter_home/Documents/scripts/browser/'
-   
-if platform.system() == 'Windows':
-    writing_path = 'c:/users/peter/Nextcloud/Documents/writing/'
-else:
-    writing_path = '/media/hd2/peter_home/Documents/writing/'
-    img_path = '/media/hd2/peter_home/Documents/writing_img/'
-
-if platform.system() == 'Windows':
-    pics_path = 'c:/users/peter/Nextcloud/misc/'
-else:
-    pics_path = '/media/hd2/peter_home/misc/'
 
 br = "<br />"
 nl = "\n"