Context Navigation

source: Validate External Links/validate_external_links.sh @ 1177

Last change on this file since 1177 was 1177, checked in by iritscen, 5 months ago
ValExtLinks now skips URLs that aren't HTTP(S) protocol. Added some error-checking on line parsing. Added my email address.
File size: 57.5 KB

Line
1	#!/bin/bash
2
3	# Validate External Links by Iritscen (iritscen@yahoo.com)
4	#
5	# Validates a list of external links in CSV format. The resulting logs are produced in three formats:
6	# - TXT (for easy diffing with an earlier log)
7	# - RTF (for reading as a local file with clickable links)
8	# - HTML (for reading as a web page)
9	# Call script with "--help" argument for documentation. Also see Read Me.rtf for critical notes.
10	#
11	# Recommended rule:
12	# \|----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----\|
13	#
14	# Table of contents (sections of script in order of appearance, not execution):
15	# • Globals
16	# • Help Output
17	# • Setup
18	# • Utility Functions
19	# • Summary Output
20	# • Initialization
21	# • Data Sourcing
22	# • Config Output
23	# • Legend Output
24	# • Main Loop
25
26	# Set separator token to newline
27	IFS="
28	"
29
30	### GLOBALS ###
31	# Settings -- these will be changed from their defaults by the arguments passed in to the script
32	LINKS_URL="" # download external link CSV from this location (can use "file://" protocol)
33	EXCEPT_URL="" # location of wiki page with a list of exceptions for NG results
34	OUTPUT_DIR="" # place reports and all other output in a folder inside this existing folder
35	RECORD_OK_LINKS=0 # record response code to the log even when it's a value in OK_CODES
36	SHOW_SLASH=0 # record issue when a slash is added to the end of a URL
37	SHOW_HTTPS=0 # record issue when "http" is upgraded to "https"
38	SHOW_YT_RD=0 # record redirection for a youtu.be URL expanding to the full URL
39	SUGGEST_SNAPSHOTS_NG=0 # query the Internet Archive for a possible snapshot URL for each NG page
40	SUGGEST_SNAPSHOTS_OK=0 # query the Internet Archive for an existing snapshot of each OK page
41	CHECK_ARCHIVE_LINKS=0 # check URLs on archive.org and archive.is
42	TAKE_PAGE_SHOT=0 # take a screenshot of each OK page
43	TIMEOUT=10 # time to wait for a response when querying a site
44	CHROME_PATH="" # path to a copy of Google Chrome that has the command-line screenshot feature
45	URL_START=1 # start at this URL in LINKS_FILE
46	URL_LIMIT=0 # if non-zero, stop at this URL in LINKS_FILE
47	UPLOAD_INFO="" # path to a file on your hard drive with the login info needed to upload a report
48
49	# Fixed strings -- see the occurrences of these variables to learn their purpose
50	AGENT="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/103.0.5060.134 Safari/537.36"
51	ARCHIVE_API="http://archive.org/wayback/available"
52	ARCHIVE_GENERIC="https://web.archive.org/web/*"
53	ARCHIVE_OK_CODES="statuscodes=200&statuscodes=203&statuscodes=206"
54	CHROME_SCREENSHOT="screenshot.png"
55	EXCEPT_FILE_NAME="exceptions.txt"
56	EXPECT_SCRIPT_NAME="val_expect_sftp.txt"
57	WIKI_CURL="https://wiki.oni2.net/Validate_External_Links/Curl_codes"
58	WIKI_HTTP="https://wiki.oni2.net/Validate_External_Links/HTTP_codes"
59	WIKI_MAIN="https://wiki.oni2.net/Validate_External_Links"
60	WIKI_ME="http://iritscen.oni2.net"
61	THIS_DIR=$(cd $(dirname $0); pwd)
62	WORKING_DIR=$(pwd)
63	WIKI_PATH="wiki.oni2.net"
64
65	# These are parallel arrays of the IDs and names of OniGalore's current namespaces
66	declare -a NS_IDS=(-2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 100 101 102 103 104 105 108 109 110 111)
67	declare -a NS_NAMES=("Media" "Special" "Main" "Talk" "User" "User_talk" "OniGalore" "OniGalore_talk" "File" "File_talk" "MediaWiki" "MediaWiki_talk" "Template" "Template_talk" "Help" "Help_talk" "Category" "Category_talk" "BSL" "BSL_talk" "OBD" "OBD_talk" "AE" "AE_talk" "Oni2" "Oni2_talk" "XML" "XML_talk")
68
69	# These arrays tell the script which suffixes at the ends of URLs represent files and which are pages.
70	# This determines whether the script tries to take a screenshot of the URL or just gets its HTTP code.
71	declare -a HTTP_FILES=(3ds 7z avi BINA blend bsl cpp csv dae dll dmg doc east exe fbx first63 flv gamedata gif jpg last32 log m4a mhm mov mp3 mp4 oni ONWC pdf png psd py rar tga TRMA txt vbs wav wmv xaf xcf xlsx xml zip)
72	declare -a HTTP_TLDS_AND_PAGES=(abstract action ars asp aspx cfm cgi com css de do full htm html it js jsp net org pgi php php3 phtml pl ru shtml stm uk x)
73
74	# These arrays tells us which HTTP response codes are OK (good), which are RD (redirections), and which
75	# are NG (no good). Pages that return OK codes will be screenshotted. Remember to update http_codes.txt
76	# if you add a new code.
77	declare -a OK_CODES=(200 401 405 406 418 501)
78	declare -a RD_CODES=(301 302 303 307 308)
79	declare -a NG_CODES=(000 400 403 404 410 429 500 502 503 504 530)
80
81	# Characters not allowed in a URL. Curly braces are sometimes used on the wiki to build a link using
82	# transcluded text, and if the transclusion fails, then the braces show up in the URL
83	ILLEGAL_CHARS="{ }"
84
85	# The shortest URL possible, used for sanity-checking some URLs: http://a.co
86	MIN_URL_LENGTH=11
87
88	# These are parallel arrays giving the prefixes that can be used in place of normal external links to
89	# some wikis and other sites; based on https://wiki.oni2.net/Special:Interwiki
90	declare -a INTERWIKI_PREFIXES=(commons metawikimedia mw wikibooks wikidata wikimedia wikinews wikiquote wikisource wikispecies wikiversity wikivoyage wikt wp)
91	declare -a INTERWIKI_DOMAINS=(commons.wikimedia.org meta.wikimedia.org mediawiki.org wikibooks.org wikidata.org wikimediafoundation.org wikinews.org wikiquote.org wikisource.org species.wikimedia.org wikiversity.org wikivoyage.org wiktionary.org wikipedia.org)
92
93	# Variables for keeping track of main loop progress and findings
94	LINK_NUM=0
95	EI_LINKS=0
96	IW_LINKS=0
97	OK_LINKS=0
98	RD_LINKS=0
99	NG_LINKS=0
100	SKIP_PARSE_FAIL=0
101	SKIP_UNK_PROT=0
102	SKIP_UNK_NS=0
103	SKIP_JS_PAGE=0
104	SKIP_BAD_URL=0
105	SKIP_NON_ASCII=0
106	SKIP_UNK_SUFFIX=0
107	SKIP_UNK_CODE=0
108	SKIP_EXPECT_NG=0
109	SKIP_EXPECT_RD=0
110	SKIP_EXPECT_EI=0
111	SKIP_EXPECT_IW=0
112	SKIP_HTTPS_UP=0
113	SKIP_SLASH_ADD=0
114	SKIP_YOUTU_BE=0
115	SKIP_ARCHIVES=0
116	FILE_LINKS=0
117	PAGE_LINKS=0
118	SKIPPED_HEADER_ROW=0
119	FINISHED_LIST="no"
120	START_RUN=0
121	END_RUN=0
122
123
124	### HELP OUTPUT ###
125	# A pseudo-man page. Here is the 80-character rule for the page text:
126	# 234567890123456789012345678901234567890123456789012345678901234567890123456789
127	function printHelp()
128	{
129	cat << EOF
130
131	NAME
132	Validate External Links
133
134	SYNOPSIS
135	validate_external_links.sh --help
136	validate_external_links.sh --links URL --output DIR [--exceptions URL]
137	[--record-ok-links] [--show-added-slashes] [--show-https-upgrades]
138	[--show-yt-redirects] [--suggest-snapshots] [--check-archive-links]
139	[--take-screenshots FILE] [--timeout NUM] [--start-url NUM]
140	[--end-url NUM] [--upload FILE]
141
142	DESCRIPTION
143	This script parses a list of external links found in the OniGalore wiki
144	(which is dumped by the Oni2.net server periodically in a particular
145	format), validates them using the Unix tool 'curl', and produces a report
146	of which links were "OK" (responded positively to an HTTP query), which
147	were "RD" (responded with a 3xx redirect code), which could be "IW"
148	(interwiki) links, which are "EI" (external internal) links and could be
149	intrawiki links, and which were "NG" (no good; a negative response to the
150	query). This report can then be automatically uploaded to the location of
151	your choice. The script can also suggest Internet Archive snapshots for
152	"NG" links, and take screenshots of "OK" links for visual verification by
153	the reader that the page in question is the one intended to be displayed.
154
155	You must pass this script the URL at which the list of links is found
156	(--links) and the path where the directory of logs should be outputted
157	(--output). All other arguments are optional.
158
159	OPTIONS
160	--help Show this page.
161	--links URL (required) URL from which to download the CSV
162	file with external links. Note that this URL can
163	be a local file if you supply a file:// path.
164	--output DIR (required) Unix path to directory in which Val
165	should place its reports.
166	--exceptions URL In order to remove links from the report which
167	Val finds an issue with but which you regard as
168	OK, list those desired exceptions on a wiki page.
169	See the sample file "exceptions.pdf" for the
170	required format of the page. Note that this URL
171	can point to a local file if you supply a path
172	beginning with "file://".
173	--record-ok-links Log a link in the report even if its response
174	code is "OK".
175	--show-added-slashes Report on redirects that simply add a '/' to the
176	end of the URL.
177	--show-https-upgrades Report on redirects that simply upgrade a
178	"http://" URL to a "https://" URL.
179	--show-yt-redirects Report on redirects that expand a youtu.be URL.
180	--suggest-snapshots-ng Query the Internet Archive for a possible
181	snapshot URL for each "NG" page.
182	--suggest-snapshots-ok Query the Internet Archive for a snapshot of each
183	"OK" page just to make sure it's available. Note
184	that this will add a tremendous amount of time to
185	the script execution because there is a rate
186	limit to the Archive API. Note that this option
187	does nothing unless you also use the
188	--record-ok-links argument.
189	--check-archive-links Check links that are already pointing to a page
190	on the Internet Archive or archive.is (AKA
191	archive.today). In theory these links should be
192	totally stable and not need validation.
193	--take-screenshots FILE Call the Google Chrome binary at this path to
194	take screenshots of each "OK" page.
195	--timeout NUM Wait this many seconds for a site to respond. The
196	default is 10. Important note: Val will attempt
197	to reach each URL three times, so the time taken
198	to ping an unresponsive site will be three times
199	this setting.
200	--start-url NUM Start at this link in the links CSV file.
201	--end-url NUM Stop at this link in the links CSV file.
202	--upload FILE Upload report using the credentials and path
203	given in this local text file. See sftp_login.txt
204	for template.
205
206	BUGS
207	The script cannot properly parse any line in the external links file
208	which contains a comma in the name of the wiki page containing a link.
209	Commas in the link itself are not an issue.
210	EOF
211	}
212
213
214	### SETUP ###
215	# If first argument is a help request, or if nothing was passed in at all, print help page and quit
216	if [ "$#" -eq 0 ] \|\| [ "$1" == "--help" ]; then
217	printHelp \| less
218	exit 0
219	fi
220
221	# Parse arguments as long as there are more arguments to process
222	while (( "$#" )); do
223	case "$1" in
224	--links ) LINKS_URL="$2"; shift 2;;
225	--exceptions ) EXCEPT_URL="$2"; shift 2;;
226	--output ) OUTPUT_DIR="$2"; shift 2;;
227	--record-ok-links ) RECORD_OK_LINKS=1; shift;;
228	--show-added-slashes ) SHOW_SLASH=1; shift;;
229	--show-https-upgrades ) SHOW_HTTPS=1; shift;;
230	--show-yt-redirects ) SHOW_YT_RD=1; shift;;
231	--suggest-snapshots-ng ) SUGGEST_SNAPSHOTS_NG=1; shift;;
232	--suggest-snapshots-ok ) SUGGEST_SNAPSHOTS_OK=1; shift;;
233	--check-archive-links ) CHECK_ARCHIVE_LINKS=1; shift;;
234	--take-screenshots ) TAKE_PAGE_SHOT=1; CHROME_PATH="$2"; shift 2;;
235	--timeout ) TIMEOUT=$2; shift 2;;
236	--start-url ) URL_START=$2; shift 2;;
237	--end-url ) URL_LIMIT=$2; shift 2;;
238	--upload ) UPLOAD_INFO=$2; shift 2;;
239	* ) echo "Invalid argument '$1' detected. Aborting."; exit 1;;
240	esac
241	done
242
243	# If the required arguments were not supplied, print help page and quit
244	if [ -z $LINKS_URL ] \|\| [ -z $OUTPUT_DIR ]; then
245	echo "Error: I did not receive one or both required arguments. Run me with the \"--help\" argument for documentation."
246	exit 2
247	fi
248
249	# If user wants screenshots, make sure path to Chrome was passed in and is valid
250	if [ $TAKE_PAGE_SHOT -eq 1 ]; then
251	if [ ! -f "$CHROME_PATH" ]; then
252	echo "Error: You need to supply a path to the Google Chrome application in order to take screenshots."
253	exit 3
254	fi
255	fi
256
257	# Check that UPLOAD_INFO exists, if this argument was supplied
258	if [ ! -z $UPLOAD_INFO ] && [ ! -f "$UPLOAD_INFO" ]; then
259	echo "Error: The file $UPLOAD_INFO supplied by the --upload argument does not appear to exist. Aborting."
260	exit 4
261	fi
262
263	# Check that OUTPUT_DIR is a directory
264	if [ ! -d "$OUTPUT_DIR" ]; then
265	echo "Error: The path $OUTPUT_DIR supplied by the --output argument does not appear to be a directory. Aborting."
266	exit 5
267	fi
268
269	# Make timestamped folder inside OUTPUT_DIR for this session's log and screenshots
270	SAFE_TIME=$(date "+%Y-%m-%d--%H-%M-%S") # for use in path names
271	NICE_TIME=$(TZ=":UTC" date "+%a, %d %b %Y %k:%M:%S GMT") # matches format of LINKS_DATE, which is extracted from the extlinks file
272	OUTPUT_FOLDER="ValExtLinks ($SAFE_TIME)"
273	OUTPUT_PATH="$OUTPUT_DIR/$OUTPUT_FOLDER"
274	SHOT_PATH="$OUTPUT_PATH/Screenshots"
275	LOG_NAME="ValExtLinks report"
276	LOG_NAME_TXT="$LOG_NAME.txt"
277	LOG_NAME_RTF="$LOG_NAME.rtf"
278	LOG_NAME_HTM="$LOG_NAME.htm"
279	LOG_PATH="$OUTPUT_PATH/$LOG_NAME"
280	LOG_PATH_TXT="$LOG_PATH.txt"
281	LOG_PATH_RTF="$LOG_PATH.rtf"
282	LOG_PATH_HTM="$LOG_PATH.htm"
283	mkdir "$OUTPUT_PATH"
284	if [ $TAKE_PAGE_SHOT -eq 1 ]; then
285	mkdir "$SHOT_PATH"
286	fi
287
288	# Check that 'mkdir' succeeded
289	if [ ! -d "$OUTPUT_PATH" ]; then
290	echo "Error: I could not create the folder \"$OUTPUT_FOLDER\" inside the directory $OUTPUT_PATH. Aborting."
291	exit 6
292	fi
293
294	# Get date on the file at LINKS_URL and print to log
295	LINKS_DATE=$(curl --silent --head $LINKS_URL \| grep "Last-Modified")
296	if [ -z "$LINKS_DATE" ]; then
297	echo "Error: I could not find the external links file at the path \"$LINKS_URL\" supplied by the --links argument. Aborting."
298	exit 7
299	fi
300	LINKS_DATE=${LINKS_DATE#Last-Modified: }
301
302
303	### UTILITY FUNCTIONS ###
304	# Writes a plain-text header to TXT log file
305	function printTXTheader()
306	{
307	valPrint t "Validate External Links report"
308	valPrint t "generated $NICE_TIME"
309	valPrint t "from data of $LINKS_DATE"
310	valPrint t "script by Iritscen (contact: $WIKI_ME)"
311	valPrint t ""
312	}
313
314	# Writes the RTF header to RTF log file, then a centered title, then sets text to left-justified
315	function printRTFheader()
316	{
317	valPrint r "{\rtf1\ansi\ansicpg1252\cocoartf1504\cocoasubrtf820
318	{\fonttbl\f0\fswiss\fcharset0 Helvetica;}
319	{\colortbl;\red255\green255\blue255;}
320	{\*\expandedcolortbl;;}
321	\margl1440\margr1440\vieww12600\viewh12100\viewkind0
322	\pard\tx720\tx1440\tx2160\tx2880\tx3600\tx4320\tx5040\tx5760\tx6480\tx7200\tx7920\tx8640\li360\pardirnatural\qc\partightenfactor0
323
324	\f0\fs28 \cf0 \b1 Validate External Links report \b0\\
325	generated $NICE_TIME\\
326	from data of $LINKS_DATE\\
327	script by Iritscen ({\field{\*\fldinst{HYPERLINK \"$WIKI_ME\"}}{\fldrslt contact}})
328	\\
329	\pard\tx720\tx1440\tx2160\tx2880\tx3600\tx4320\tx5040\tx5760\tx6480\tx7200\tx7920\tx8640\li360\pardirnatural\partightenfactor0
330	\cf0 "
331	}
332
333	# Closes the RTF markup of the RTF log file
334	function printRTFfooter()
335	{
336	valPrint r "}"
337	}
338
339	# Writes the HTML header to HTML log file
340	function printHTMheader()
341	{
342	valPrint h "<html>
343	<head>
344	<title>Validate External Links report</title>
345	</head>
346	<body>
347	<h2>Validate External Links report</h2>
348	<h3>generated $NICE_TIME<br />
349	from data of $LINKS_DATE<br />
350	script by Iritscen (<a href=\"$WIKI_ME\" target=\"_blank\">contact</a>)</h3>"
351	}
352
353	# Closes the HTML markup of the HTML log file
354	function printHTMfooter()
355	{
356	valPrint h "</body>
357	</html>"
358	}
359
360	# The central logging function. The first parameter is a string composed of one or more characters that
361	# indicate which output to use: 'c' means console, 't' means the TXT log, 'r' means the RTF log, and
362	# 'h' means the HTML log. 'n' means "Don't print a newline at the end of the line." 's' means "Print an
363	# extra newline at the end." 'w' means "Don't pass console output through 'fmt'" ("fmt" fits the output
364	# to an 80-column CLI but can break special formatting and the 'n' option).
365	function valPrint()
366	{
367	if [[ "$1" == c ]]; then
368	if [[ "$1" == n ]]; then
369	echo -n "$2"
370	elif [[ "$1" == w ]]; then
371	echo "$2"
372	elif [[ "$1" == s ]]; then
373	echo -e "$2\n"
374	else
375	echo "$2" \| fmt -w 80
376	fi
377	fi
378	if [[ "$1" == t ]]; then
379	if [[ "$1" == n ]]; then
380	echo -n "$2" >> "$LOG_PATH_TXT"
381	elif [[ "$1" == s ]]; then
382	echo -e "$2\n" >> "$LOG_PATH_TXT"
383	else
384	echo "$2" >> "$LOG_PATH_TXT"
385	fi
386	fi
387	if [[ "$1" == r ]]; then
388	if [[ "$1" == n ]]; then
389	echo "$2" >> "$LOG_PATH_RTF"
390	elif [[ "$1" == s ]]; then
391	echo "$2\line\line" >> "$LOG_PATH_RTF"
392	else
393	echo "$2\line" >> "$LOG_PATH_RTF"
394	fi
395	fi
396	if [[ "$1" == h ]]; then
397	if [[ "$1" == s ]]; then
398	echo "$2<tr><td> </td></tr>" >> "$LOG_PATH_HTM"
399	elif [[ "$1" == n ]]; then
400	echo "$2" >> "$LOG_PATH_HTM"
401	else
402	echo "$2<br />" >> "$LOG_PATH_HTM"
403	fi
404	fi
405	}
406
407	# Pluralize the string in parameter 1 if the number in parameter 2 is not 1
408	function pluralCheckNoun()
409	{
410	if [ $2 -ne 1 ]; then
411	if [[ $1 =~ x$ ]]; then
412	echo $1es
413	else
414	echo $1s
415	fi
416	else
417	echo $1
418	fi
419	}
420
421	# Output "is" if parameter 1 is 1, otherwise "are"
422	function pluralCheckIs()
423	{
424	if [ $1 -ne 1 ]; then
425	echo "are"
426	else
427	echo "is"
428	fi
429	}
430
431	# Output "was" if parameter 1 is 1, otherwise "were"
432	function pluralCheckWas()
433	{
434	if [ $1 -ne 1 ]; then
435	echo "were"
436	else
437	echo "was"
438	fi
439	}
440
441	# Output "a " if parameter 1 is 1, otherwise nothing
442	function pluralCheckA()
443	{
444	if [ $1 -eq 1 ]; then
445	echo "a "
446	fi
447	}
448
449	# Output "an " if parameter 1 is 1, otherwise nothing
450	function pluralCheckAn()
451	{
452	if [ $1 -eq 1 ]; then
453	echo "an "
454	fi
455	}
456
457	# Upload the reports using info specified in the --upload argument. ONLY USE "valPrint c" here, as the
458	# reports being saved to disk have already been closed.
459	function uploadReport()
460	{
461	valPrint c "Uploading reports..."
462
463	SCRIPT_PATH="$THIS_DIR/$EXPECT_SCRIPT_NAME"
464	SFTP_USER_NAME_MARKER="user:"
465	SFTP_PASSWORD_MARKER="pw:"
466	SFTP_PORT_MARKER="port:"
467	SFTP_PATH_MARKER="path:"
468	SFTP_USER_NAME=$(grep $SFTP_USER_NAME_MARKER $UPLOAD_INFO)
469	SFTP_USER_NAME=${SFTP_USER_NAME#$SFTP_USER_NAME_MARKER}
470	SFTP_PASSWORD=$(grep $SFTP_PASSWORD_MARKER $UPLOAD_INFO)
471	SFTP_PASSWORD=${SFTP_PASSWORD#$SFTP_PASSWORD_MARKER}
472	SFTP_PORT=$(grep $SFTP_PORT_MARKER $UPLOAD_INFO)
473	SFTP_PORT=${SFTP_PORT#$SFTP_PORT_MARKER}
474	SFTP_PATH=$(grep $SFTP_PATH_MARKER $UPLOAD_INFO)
475	SFTP_PATH=${SFTP_PATH#$SFTP_PATH_MARKER}
476
477	for SUFFIX in htm rtf txt; do
478	expect "$SCRIPT_PATH" "$LOG_PATH.$SUFFIX" $SFTP_USER_NAME $SFTP_PASSWORD $SFTP_PORT $SFTP_PATH "$LOG_NAME.$SUFFIX"
479
480	if [ "$?" -ne 0 ]; then
481	valPrint c "Error $? occurred when attempting to upload $LOG_NAME.$SUFFIX!"
482	else
483	valPrint c "Report in `echo $SUFFIX \| tr [:lower:] [:upper:]` format was uploaded."
484	fi
485	done
486	}
487
488	# Prints session summary when script is done
489	function wrapupAndExit()
490	{
491	# Get off progress line on console, drop down a line from last link in log, and close HTML table
492	valPrint ctr ""
493	valPrint h "</table><br />"
494
495	# If we didn't finish processing the last URL, then the iterator is one too high
496	if [ $FINISHED_LIST != "yes" ]; then
497	let LINK_NUM-=1
498	if [ $FINISHED_LIST == "no" ]; then
499	valPrint ctrh "The session was canceled by the user."
500	fi
501	fi
502
503	# Generate string with elapsed time
504	END_RUN=$(date +%s)
505	ELAPSED=$(echo $(($END_RUN - $START_RUN)) \| awk '{printf "%d min. %d sec. elapsed", int($1/60), int($1%60)}')
506
507	# Do some math on results of session
508	LINKS_PROCESSED=$((LINK_NUM-URL_START+1))
509	TRIVIAL_RDS=$((SKIP_SLASH_ADD+SKIP_HTTPS_UP+SKIP_YOUTU_BE))
510	LINK_ERRORS=$((SKIP_PARSE_FAIL+SKIP_UNK_PROT+SKIP_UNK_NS+SKIP_JS_PAGE+SKIP_BAD_URL+SKIP_NON_ASCII+SKIP_UNK_SUFFIX+SKIP_UNK_CODE))
511	LINKS_EXCEPTED=$((SKIP_EXPECT_NG+SKIP_EXPECT_RD+SKIP_EXPECT_EI+SKIP_EXPECT_IW))
512	LINK_PROBLEMS_TOTAL=$((NG_LINKS+RD_LINKS+EI_LINKS+IW_LINKS))
513	LINK_PROBLEMS_NG=$((NG_LINKS-SKIP_EXPECT_NG))
514	LINK_PROBLEMS_RD=$((RD_LINKS-SKIP_EXPECT_RD))
515	LINK_PROBLEMS_EI=$((EI_LINKS-SKIP_EXPECT_EI))
516	LINK_PROBLEMS_IW=$((IW_LINKS-SKIP_EXPECT_IW))
517	LINK_PROBLEMS_NET=$((LINK_PROBLEMS_NG+LINK_PROBLEMS_RD+LINK_PROBLEMS_EI+LINK_PROBLEMS_IW))
518
519	# Print something in the Links section if no link issues were printed
520	if [ $LINK_PROBLEMS_NET -eq 0 ]; then
521	valPrint h "<i>No link problems to report! See the <a href=\"$LOG_NAME_RTF\" target=\"_blank\">RTF</a> or <a href=\"$LOG_NAME_TXT\" target=\"_blank\">TXT</a> report for a list of links with issues that were not reported.</i>"
522	fi
523	if [ $LINK_PROBLEMS_TOTAL -eq 0 ]; then
524	valPrint t "No link problems to report!"
525	valPrint r "\i1 No link problems to report! \i0"
526	fi
527
528	## SUMMARY OUTPUT ##
529	valPrint ct "Summary ($ELAPSED):"
530	valPrint r "\b1 Summary \b0 ($ELAPSED)"
531	valPrint hn "<h3><span id=\"summary\">Summary ($ELAPSED)</span></h3>"
532	valPrint ctrh "I finished processing $LINKS_PROCESSED of $LINK_COUNT $(pluralCheckNoun link $LINK_COUNT) (there $(pluralCheckWas $FILE_LINKS) $FILE_LINKS file $(pluralCheckNoun link $FILE_LINKS) and $PAGE_LINKS page $(pluralCheckNoun link $PAGE_LINKS))."
533
534	# Print processed link totals
535	if [ $LINKS_PROCESSED -gt 0 ]; then valPrint ctrh "$LINKS_PROCESSED processed $(pluralCheckNoun link $LINKS_PROCESSED):"; fi
536	if [ $LINK_ERRORS -gt 0 ]; then valPrint ctrh "- $LINK_ERRORS $(pluralCheckNoun link $LINK_ERRORS) could not be processed"; fi
537	if [ $SKIP_ARCHIVES -gt 0 ]; then valPrint ctrh "- $SKIP_ARCHIVES archive.org/archive.is $(pluralCheckNoun link $SKIP_ARCHIVES) were not checked"; fi
538	if [ $LINK_PROBLEMS_TOTAL -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_TOTAL processed $(pluralCheckNoun link $LINK_PROBLEMS_TOTAL) had $(pluralCheckAn $LINK_PROBLEMS_TOTAL)$(pluralCheckNoun issue $LINK_PROBLEMS_TOTAL)"; fi
539	if [ $LINKS_EXCEPTED -gt 0 ]; then valPrint ctr " (excepted $LINKS_EXCEPTED link $(pluralCheckNoun issue $LINKS_EXCEPTED) from report)"; valPrint h "  (excepted $LINKS_EXCEPTED $(pluralCheckNoun link $LINKS_EXCEPTED) from report)"; fi
540	if [ $OK_LINKS -gt 0 ]; then valPrint ctrh "- $OK_LINKS processed $(pluralCheckNoun link $OK_LINKS) $(pluralCheckWas $OK_LINKS) OK"; fi
541	if [ $TRIVIAL_RDS -gt 0 ]; then valPrint ctr " (counted $TRIVIAL_RDS trivial $(pluralCheckNoun redirection $TRIVIAL_RDS) as OK)"; valPrint h "  (counted $TRIVIAL_RDS trivial $(pluralCheckNoun redirection $TRIVIAL_RDS) as OK)"; fi
542
543	# Print errored link totals
544	if [ $LINK_ERRORS -gt 0 ]; then
545	valPrint c "$LINK_ERRORS link $(pluralCheckNoun error $LINK_ERRORS) (see RTF or TXT report for specific links):"
546	valPrint h "$LINK_ERRORS link $(pluralCheckNoun error $LINK_ERRORS) (see <a href=\"$LOG_NAME_RTF\" target=\"_blank\">RTF</a> or <a href=\"$LOG_NAME_TXT\" target=\"_blank\">TXT</a> report for specific links):"
547	valPrint rt "$LINK_ERRORS link $(pluralCheckNoun error $LINK_ERRORS):"
548	fi
549	if [ $SKIP_PARSE_FAIL -gt 0 ]; then valPrint ctrh "- $SKIP_PARSE_FAIL line-parsing $(pluralCheckNoun failure $SKIP_PARSE_FAIL)"; fi
550	if [ $SKIP_UNK_PROT -gt 0 ]; then valPrint ctrh "- $SKIP_UNK_PROT unknown $(pluralCheckNoun protocol $SKIP_UNK_PROT)"; fi
551	if [ $SKIP_UNK_NS -gt 0 ]; then valPrint ctrh "- $SKIP_UNK_NS missing/unknown $(pluralCheckNoun namespace $SKIP_UNK_NS)"; fi
552	if [ $SKIP_JS_PAGE -gt 0 ]; then valPrint ctrh "- $SKIP_JS_PAGE $(pluralCheckNoun link $SKIP_JS_PAGE) on $(pluralCheckA $SKIP_JS_PAGE)JavaScript $(pluralCheckNoun page $SKIP_JS_PAGE)"; fi
553	if [ $SKIP_BAD_URL -gt 0 ]; then valPrint ctrh "- $SKIP_BAD_URL illegal $(pluralCheckNoun URL $SKIP_BAD_URL)"; fi
554	if [ $SKIP_NON_ASCII -gt 0 ]; then valPrint ctrh "- $SKIP_NON_ASCII non-ASCII $(pluralCheckNoun URL $SKIP_NON_ASCII)"; fi
555	if [ $SKIP_UNK_SUFFIX -gt 0 ]; then valPrint ctrh "- $SKIP_UNK_SUFFIX unknown URL $(pluralCheckNoun suffix $SKIP_UNK_SUFFIX)"; fi
556	if [ $SKIP_UNK_CODE -gt 0 ]; then valPrint ctrh "- $SKIP_UNK_CODE unknown response $(pluralCheckNoun code $SKIP_UNK_CODE)"; fi
557
558	# Print excepted link totals
559	if [ $LINKS_EXCEPTED -gt 0 ]; then
560	valPrint c "$LINKS_EXCEPTED link $(pluralCheckNoun problem $LINKS_EXCEPTED) excepted (see RTF or TXT report for specific links):"
561	valPrint h "$LINKS_EXCEPTED link $(pluralCheckNoun problem $LINKS_EXCEPTED) excepted (see <a href=\"$LOG_NAME_RTF\" target=\"_blank\">RTF</a> or <a href=\"$LOG_NAME_TXT\" target=\"_blank\">TXT</a> report for specific links):"
562	valPrint rt "$LINKS_EXCEPTED link $(pluralCheckNoun problem $LINKS_EXCEPTED) excepted:"
563	fi
564	if [ $SKIP_EXPECT_NG -gt 0 ]; then valPrint ctrh "- $SKIP_EXPECT_NG/$NG_LINKS NG $(pluralCheckNoun link $NG_LINKS)"; fi
565	if [ $SKIP_EXPECT_RD -gt 0 ]; then valPrint ctrh "- $SKIP_EXPECT_RD/$RD_LINKS $(pluralCheckNoun redirection $RD_LINKS)"; fi
566	if [ $SKIP_EXPECT_EI -gt 0 ]; then valPrint ctrh "- $SKIP_EXPECT_EI/$EI_LINKS external internal $(pluralCheckNoun link $EI_LINKS)"; fi
567	if [ $SKIP_EXPECT_IW -gt 0 ]; then valPrint ctrh "- $SKIP_EXPECT_IW/$IW_LINKS potential intrawiki $(pluralCheckNoun link $IW_LINKS)"; fi
568
569	# Perform exceptions audit
570	EXCEPTION_ISSUES=0
571	valPrint ctrh "Exceptions list audit:"
572	for ((i = 0; i < ${#EXCEPT_ARRAY[@]}; ++i)); do
573	EXCEPT_LINE="${EXCEPT_ARRAY[$i]}"
574	EXCEPT_LINE=$(echo "$EXCEPT_LINE" \| sed 's/\&/\&/g') # copied from exception-matching code
575
576	if [ ${EXCEPT_FOUND[$i]} -eq 0 ]; then
577	EXCEPT_URL="${EXCEPT_LINE#*,}"
578	EXCEPT_URL="${EXCEPT_URL%,*}"
579	EXCEPT_PAGE="${EXCEPT_LINE##*,}"
580	EXCEPT_PAGE="${EXCEPT_PAGE%% *}"
581	if [ "$EXCEPT_PAGE" == "*" ]; then
582	valPrint tr "- The link '$EXCEPT_URL' did not return an error or is not on any page."
583	else
584	valPrint tr "- The link '$EXCEPT_URL' did not return an error or is not on page '$EXCEPT_PAGE'."
585	fi
586	let EXCEPTION_ISSUES+=1
587	elif [ ${EXCEPT_USED[$i]} -eq 0 ]; then
588	EXCEPT_URL="${EXCEPT_LINE#*,}"
589	EXCEPT_URL="${EXCEPT_URL%,*}"
590	EXCEPT_CODE=${EXCEPT_LINE%%,*}
591	valPrint tr "- The link '$EXCEPT_URL' did not return error code $EXCEPT_CODE."
592	let EXCEPTION_ISSUES+=1
593	fi
594	done
595	if [ $EXCEPTION_ISSUES -eq 0 ]; then
596	valPrint ctrh "- No issues found."
597	else
598	valPrint c "$EXCEPTION_ISSUES exception list $(pluralCheckNoun issue $EXCEPTION_ISSUES) detected (see RTF or TXT report for details)."
599	valPrint h "$EXCEPTION_ISSUES exception list $(pluralCheckNoun issue $EXCEPTION_ISSUES) detected (see <a href=\"$LOG_NAME_RTF\" target=\"_blank\">RTF</a> or <a href=\"$LOG_NAME_TXT\" target=\"_blank\">TXT</a> report for details)."
600	fi
601
602	# Print checked link totals
603	if [ $LINK_PROBLEMS_NET -gt 0 ]; then valPrint ctrh "$LINK_PROBLEMS_NET link $(pluralCheckNoun issue $LINK_PROBLEMS_NET):"; fi
604	if [ $LINK_PROBLEMS_NG -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_NG NG $(pluralCheckNoun link $LINK_PROBLEMS_NG)"; fi
605	if [ $LINK_PROBLEMS_RD -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_RD $(pluralCheckNoun redirection $LINK_PROBLEMS_RD)"; fi
606	if [ $LINK_PROBLEMS_EI -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_EI $(pluralCheckNoun link $LINK_PROBLEMS_EI) that could be intrawiki"; fi
607	if [ $LINK_PROBLEMS_IW -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_IW $(pluralCheckNoun link $LINK_PROBLEMS_IW) that could be interwiki"; fi
608
609	# Close the log files' markup
610	valPrint trh "ValExtLinks says goodbye."
611	printRTFfooter
612	printHTMfooter
613
614	# Upload report if this was requested
615	if [ ! -z $UPLOAD_INFO ] && [ $FINISHED_LIST != "no" ]; then
616	uploadReport
617	fi
618
619	# Really quit now
620	valPrint c "ValExtLinks says goodbye."
621	exit 0
622	}
623	trap wrapupAndExit INT
624
625
626	### INITIALIZATION ###
627	# Print opening message to console and log files
628	valPrint cw "$(tput bold)--Validate External Links--$(tput sgr0)"
629	printTXTheader
630	printRTFheader
631	printHTMheader
632
633	## DATA SOURCING ##
634	valPrint t "Startup:"
635	valPrint r "\b1 Startup \b0"
636	valPrint hn "<h3>Startup</h3>"
637
638	# Attempt to download file at LINKS_URL, then check that it succeeded
639	valPrint cwtrhn "Downloading list of external links from $LINKS_URL..."
640	LINKS_FILE_NAME=$(echo "$LINKS_URL" \| sed 's/.*\///')
641	LINKS_FILE="$OUTPUT_PATH/$LINKS_FILE_NAME"
642	curl --silent -o "$LINKS_FILE" $LINKS_URL
643	if [ ! -f "$LINKS_FILE" ]; then
644	echo -e "\nThe download of $LINKS_URL appears to have failed. Aborting."
645	wrapupAndExit
646	else
647	valPrint ctrh " success."
648	fi
649
650	# Attempt to download file at EXCEPT_URL, then check that it succeeded
651	if [ ! -z $EXCEPT_URL ]; then
652	valPrint cwtrhn "Downloading list of reporting exceptions from $EXCEPT_URL..."
653	EXCEPT_DATA=$(curl --silent $EXCEPT_URL)
654	if [ -z "$EXCEPT_DATA" ]; then
655	echo -e "\nThe download of the exceptions data from '$EXCEPT_URL' appears to have failed. Aborting."
656	wrapupAndExit
657	else
658	valPrint ctrh " success."
659	fi
660	EXCEPT_DATA=${EXCEPT_DATA%END LIST*}
661	EXCEPT_DATA=${EXCEPT_DATA#*BEGIN LIST}
662	EXCEPT_FILE="$OUTPUT_PATH/$EXCEPT_FILE_NAME"
663
664	# Store on disk for debugging purposes
665	echo "$EXCEPT_DATA" > "$EXCEPT_FILE"
666
667	# Transfer to array for easy searching later
668	declare -a EXCEPT_ARRAY=($(echo "$EXCEPT_DATA"))
669
670	# Create parallel arrays for marking which exceptions get used later
671	declare -a EXCEPT_USED=()
672	declare -a EXCEPT_FOUND=()
673	for ((i = 0; i < ${#EXCEPT_ARRAY[@]}; ++i)); do
674	EXCEPT_USED+=(0)
675	EXCEPT_FOUND+=(0)
676	done
677	fi
678
679	# Pipe 'cat' to 'wc' because passing LINKS_FILE to 'wc' would print the file name after the line count
680	LINK_COUNT_STRING=$(cat "$LINKS_FILE" \| wc -l)
681
682	# Number of URLs is number of lines minus one (first line is column header row for the CSV)
683	LINK_COUNT=$(echo "${LINK_COUNT_STRING}" \| tr -d '[:space:]')
684	let LINK_COUNT-=1
685	valPrint ctrh "Found $LINK_COUNT links to process."
686	valPrint trh ""
687
688	## CONFIG OUTPUT ##
689	valPrint t "Config:"
690	valPrint r "\b1 Config \b0"
691	valPrint hn "<h3>Config</h3>"
692
693	valPrint ctrhn "Links to consider: "
694	if [ $URL_LIMIT -ne 0 ] && [ $URL_LIMIT -lt $LINK_COUNT ]; then
695	valPrint ctrh "$((URL_LIMIT-URL_START+1)), from link $URL_START to $URL_LIMIT"
696	elif [ $URL_START -ne 1 ]; then
697	valPrint ctrh "$((LINK_COUNT-URL_START+1)), from link $URL_START to $LINK_COUNT"
698	else
699	valPrint ctrh "$LINK_COUNT"
700	fi
701
702	valPrint ctrh "Site query timeout: $TIMEOUT seconds"
703
704	valPrint ctrhn "Show OK links: "
705	if [ $RECORD_OK_LINKS -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
706
707	valPrint ctrhn "Take screenshots: "
708	if [ $TAKE_PAGE_SHOT -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
709
710	valPrint ctrhn "Suggest archive.org snapshots for NG pages: "
711	if [ $SUGGEST_SNAPSHOTS_NG -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
712
713	valPrint ctrhn "Suggest archive.org snapshots for OK pages: "
714	if [ $SUGGEST_SNAPSHOTS_OK -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
715
716	valPrint ctrhn "Ignore slash-adding redirects: "
717	if [ $SHOW_SLASH -eq 1 ]; then valPrint ctrh "No"; else valPrint ctrh "Yes"; fi
718
719	valPrint ctrhn "Ignore HTTPS-upgrading redirects: "
720	if [ $SHOW_HTTPS -eq 1 ]; then valPrint ctrh "No"; else valPrint ctrh "Yes"; fi
721
722	valPrint ctrhn "Ignore youtu.be redirects: "
723	if [ $SHOW_YT_RD -eq 1 ]; then valPrint ctrh "No"; else valPrint ctrh "Yes"; fi
724
725	valPrint ctrhn "Check archive.org and archive.is links: "
726	if [ $CHECK_ARCHIVE_LINKS -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
727
728	valPrint tr "A summary of my findings will be found at the bottom of the report."
729	valPrint h "A summary of my findings will be found at the <a href=\"#summary\">bottom</a> of the report."
730	valPrint trh ""
731
732	## LEGEND OUTPUT ##
733	valPrint t "Legend:"
734	valPrint r "\b1 Legend \b0"
735	valPrint hn "<h3>Legend</h3>"
736	valPrint t "(For guidance in fixing these links, see $WIKI_MAIN. The exceptions list is at $EXCEPT_URL.)"
737	valPrint r "(For guidance in fixing these links, see {\field{\\fldinst{HYPERLINK \"$WIKI_MAIN\"}}{\fldrslt here}}. The exceptions list is {\field{\\fldinst{HYPERLINK \"$EXCEPT_URL\"}}{\fldrslt here}}.)"
738	valPrint h "(For guidance in fixing these links, see <a href=\"$WIKI_MAIN\" target=\"_blank\">here</a>. The exceptions list is <a href=\"$EXCEPT_URL\" target=\"_blank\">here</a>.)"
739	valPrint trh "OK = URL seems to be working"
740	valPrint trh "NG = URL no longer seems to work"
741	valPrint trh "RD = URL is redirecting to this new URL"
742	valPrint trh "EI = URL is an external link to an internal page and should be converted to an intrawiki link using the suggested markup"
743	valPrint trh "IW = URL is an external link to a fellow wiki and should be converted to an interwiki link using the suggested markup"
744	valPrint t "(xxx) = Unix tool 'curl' obtained this HTTP response status code (see here for code reference: $WIKI_HTTP)"
745	valPrint r "(xxx) = Unix tool 'curl' obtained this HTTP response status code (see {\field{\*\fldinst{HYPERLINK \"$WIKI_HTTP\"}}{\fldrslt here}} for code reference)"
746	valPrint h "(xxx) = Unix tool 'curl' obtained this HTTP response status code (see <a href=\"$WIKI_HTTP\" target=\"_blank\">here</a> for code reference)"
747	valPrint t "(000-xx) = 'curl' did not get an HTTP response code, but returned this exit code (see here for code reference: $WIKI_CURL)"
748	valPrint r "(000-xx) = 'curl' did not get an HTTP response code, but returned this exit code (see {\field{\*\fldinst{HYPERLINK \"$WIKI_CURL\"}}{\fldrslt here}} for code reference)"
749	valPrint h "(000-xx) = 'curl' did not get an HTTP response code, but returned this exit code (see <a href=\"$WIKI_CURL\" target=\"_blank\">here</a> for code reference)"
750	valPrint trh "IA suggests = Last available snapshot returned by the Internet Archive"
751	valPrint trh "Try browsing = The Archive failed to return a snapshot URL, so check for a snapshot manually using this link"
752	valPrint trh ""
753
754
755	### MAIN LOOP ###
756	valPrint t "Links:"
757	valPrint r "\b1 Links \b0"
758	valPrint hn "<h3>Links</h3>"
759	START_RUN=$(date +%s)
760	# Process each line of the .csv in LINKS_FILE
761	for LINE in `cat "$LINKS_FILE"`; do
762	START_LINK=$(date +%s)
763	let LINK_NUM+=1
764
765	# First line is the column header row for the CSV, so let's verify that the format hasn't changed
766	if [ $SKIPPED_HEADER_ROW -eq 0 ]; then
767	if [ $LINE == "namespace,title,target" ]; then
768	SKIPPED_HEADER_ROW=1
769	LINK_NUM=0 # this line is not a link, so reset the link counter
770	valPrint hn "<table>"
771	continue
772	else
773	valPrint ctrh "Did not find expected header row in $LINKS_FILE. Aborting."
774	wrapupAndExit
775	fi
776	fi
777
778	# Skip this link if we are not at URL_START yet
779	if [ $LINK_NUM -lt $URL_START ]; then
780	continue
781	fi
782
783	# Stop if we are at the limit declared for testing purposes
784	if [ $URL_LIMIT -gt 0 ] && [ $LINK_NUM -gt $URL_LIMIT ]; then
785	FINISHED_LIST="limit"
786	wrapupAndExit
787	fi
788
789	# Parse line into namespace ID number, containing wiki page, and external link URL
790	NS_ID=${LINE%%,*}
791	PAGE_NAME=${LINE#$NS_ID,}
792	PAGE_NAME=${PAGE_NAME%%,*} # a comma in the page name will break this
793	URL=${LINE#$NS_ID,$PAGE_NAME,} # commas can be in this
794	if [ -z "$NS_ID" ] \|\| [ -z "$PAGE_NAME" ] \|\| [ -z "$URL" ]; then
795	valPrint trs "Skipping line $LINK_NUM ('$LINE') because the namespace, wiki page or link URL could not be read."
796	let SKIP_PARSE_FAIL+=1
797	continue
798	fi
799
800	# Skip any link that isn't "http://" or "https://"
801	if [[ ! $URL =~ ^http* ]]; then
802	valPrint trs "Skipping line $LINK_NUM ('$LINE') because the protocol isn't 'http://' or 'https://'."
803	let SKIP_UNK_PROT+=1
804	continue
805	fi
806
807	# Print progress to screen
808	if [ $LINK_NUM -gt 1 ]; then
809	printf "\e[1A\n" # erase previous progress message so that new one appears in its place
810	fi
811	valPrint cn "Evaluating URL $LINK_NUM/$LINK_COUNT..."
812
813	# Find namespace number in NS_IDS and use it to look up namespace's name in NS_NAMES
814	NS_NAME=""
815	a=0
816	while [ "x${NS_IDS[$a]}" != "x" ]; do # once this evaluates to "x", the array is done
817	if [ $NS_ID == "NULL" ]; then
818	break
819	elif [ $NS_ID -eq ${NS_IDS[$a]} ]; then
820	NS_NAME="${NS_NAMES[$a]}"
821	break
822	fi
823	let a+=1
824	done
825	if [ "$NS_NAME" == "" ]; then
826	if [ $NS_ID == "NULL" ]; then
827	valPrint trs "Skipping line $LINK_NUM ('$LINE') because the namespace (and probably the page too) is 'NULL'. Probably the link is no longer in existence on the wiki."
828	else
829	valPrint trs "Skipping line $LINK_NUM ('$LINE') because I could not find a name for namespace ID $NS_ID."
830	fi
831	let SKIP_UNK_NS+=1
832	let PAGE_LINKS+=1
833	continue
834	fi
835
836	# Build longer wiki page URLs from namespace and page names
837	FULL_PAGE_PATH=https://$WIKI_PATH/$NS_NAME:$PAGE_NAME
838	LOCAL_PAGE_PATH=$NS_NAME:$PAGE_NAME
839	# Namespace "Main:" cannot be a part of the path; it's an implicit namespace, and naming it
840	# explicitly breaks the link
841	if [ $NS_ID -eq 0 ]; then
842	FULL_PAGE_PATH=https://$WIKI_PATH/$PAGE_NAME
843	LOCAL_PAGE_PATH=$PAGE_NAME
844	fi
845
846	# We don't want to consider wiki pages ending in .js, as the MW parser cannot reliably isolate URLS
847	# in JavaScript code, so it returns erroneous links
848	PAGE_NAME_SUFFIX=$(echo $PAGE_NAME \| sed 's/.*\.//')
849	if [ $PAGE_NAME_SUFFIX == "js" ]; then
850	valPrint trs "Skipping URL '${LINE#$NS_ID,$PAGE_NAME,}' on line $LINK_NUM because it was found on JavaScript page '$LOCAL_PAGE_PATH'."
851	let SKIP_JS_PAGE+=1
852	let PAGE_LINKS+=1
853	continue
854	fi
855
856	# Scan for illegal characters
857	if [[ $URL == [$ILLEGAL_CHARS] ]]; then
858	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because it contains characters illegal in a URL."
859	let SKIP_BAD_URL+=1
860	let PAGE_LINKS+=1
861	continue
862	fi
863
864	# If we're skipping archive links, see if this is one
865	if [ $CHECK_ARCHIVE_LINKS -eq 0 ] && [[ ( $URL == web.archive.org \|\| $URL == archive.is ) ]]; then
866	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to check archive links."
867	let SKIP_ARCHIVES+=1
868	let PAGE_LINKS+=1
869	continue
870	fi
871
872	# Now we need to know if the URL is for a file or a web page. First step is to determine if the
873	# URL ends in a suffix
874	HAS_SUFFIX=0
875
876	# If the URL ends in something like ".php?foo=bar", strip everything from the '?' onward
877	CLEAN_URL=${URL%%\?*}
878
879	# If the URL ends in something like "#section_15", strip everything from the '#' onward
880	CLEAN_URL=${CLEAN_URL%%\#*}
881
882	# 'sed' cannot handle Unicode in my Bash shell, so skip non-ASCII URL and make reader check it
883	if [[ $CLEAN_URL == [![:ascii:]] ]]; then
884	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I cannot handle non-ASCII characters."
885	let SKIP_NON_ASCII+=1
886	let PAGE_LINKS+=1
887	continue
888	fi
889
890	# Isolate the characters after the last period and after the last slash
891	POST_DOT=$(echo "$CLEAN_URL" \| sed 's/.*\.//')
892	POST_SLASH=$(echo "$CLEAN_URL" \| sed 's/.*\///')
893
894	# If the last period comes after the last slash, then the URL ends in a suffix
895	POST_DOT_LENGTH=$(echo \| awk -v input=$POST_DOT '{print length(input)}')
896	POST_SLASH_LENGTH=$(echo \| awk -v input=$POST_SLASH '{print length(input)}')
897	if [ $POST_DOT_LENGTH -lt $POST_SLASH_LENGTH ]; then
898	HAS_SUFFIX=1
899	else
900	HAS_SUFFIX=0
901	fi
902
903	# Now we need to know if the URL is for a file. If the URL ends in a suffix, compare it against the
904	# known URL endings supplied in HTTP_FILES and HTTP_TLDS_AND_PAGES
905	IS_FILE=-1
906	if [ $HAS_SUFFIX -eq 0 ]; then
907	IS_FILE=0
908	else
909	# Turn off case sensitivity while we compare suffixes
910	shopt -s nocasematch
911
912	# Special case: URLs ending in something like "productID.304297400" are pages, not files, so if
913	# the URL's suffix is all numbers, we are looking at the end of a web page URL
914	if [[ $POST_DOT =~ ^-?[0-9]+$ ]]; then
915	IS_FILE=0
916	fi
917
918	# Special case: URLs ending in a parens, e.g. "ms537113(v=vs.85)", are pages
919	if [[ $POST_DOT =~ ^.*[\(\)]$ ]]; then
920	IS_FILE=0
921	fi
922
923	# Special case: URLs containing a '%', e.g. "10.1007%2FBF00329055", are pages
924	if [[ $POST_DOT == % ]]; then
925	IS_FILE=0
926	fi
927
928	# If we did not identify this URL as a web page above, we need to compare the suffix against known
929	# file extensions
930	if [ $IS_FILE -eq -1 ]; then
931	for EXTENSION in "${HTTP_FILES[@]}"; do
932	if [[ $EXTENSION == $POST_DOT ]]; then # double brackets respect the nocasematch setting
933	IS_FILE=1
934	break
935	fi
936	done
937	fi
938
939	# If we did not identify this URL as a file above, we need to compare the suffix against known
940	# pages and TLDs in order to rule out that this is an unknown file suffix or TLD that the user
941	# needs to add to the HTTP_FILES or HTTP_TLDS_AND_PAGES array
942	if [ $IS_FILE -eq -1 ]; then
943	for PAGE_OR_TLD in "${HTTP_TLDS_AND_PAGES[@]}"; do
944	if [[ $PAGE_OR_TLD == $POST_DOT ]]; then
945	IS_FILE=0
946	break
947	fi
948	done
949	fi
950
951	# Turn case sensitivity back on in Bash
952	shopt -u nocasematch
953	fi
954
955	# If this suffix escaped identification as either a file, page or TLD, inform the reader
956	STR_TYPE=""
957	if [ $IS_FILE -eq -1 ]; then
958	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I encountered the unknown URL suffix '$POST_DOT'. Please add this suffix to the appropriate array in this script, HTTP_FILES or HTTP_TLDS_AND_PAGES."
959	let SKIP_UNK_SUFFIX+=1
960	continue
961	elif [ $IS_FILE -eq 1 ]; then
962	STR_TYPE="file"
963	let FILE_LINKS+=1
964	else
965	STR_TYPE="page"
966	let PAGE_LINKS+=1
967	fi
968
969	# Get response code using 'curl' to see if this link is valid; the --insecure option avoids an
970	# issue with sites that require HTTPS
971	CURL_CODE=$(curl -o /dev/null --silent --insecure --compressed --head --user-agent '$AGENT' --max-time $TIMEOUT --retry 2 --write-out '%{http_code}\n' $URL)
972	CURL_ERR=$(echo $?)
973	CURL_RESULT=$CURL_CODE
974
975	# Tack 'curl' exit code onto result string if 'curl' returned "000" (no HTTP response)
976	if [ $CURL_CODE == "000" ]; then
977	CURL_RESULT="$CURL_RESULT-$CURL_ERR"
978	fi
979
980	# Begin to determine our status code for this URL (EI, IW, OK, RD, or NG)
981	STATUS="??"
982	NEW_URL=""
983	INTERWIKI_INDEX=-1
984
985	# First make sure that this isn't an "external internal" link to our own wiki that can be replaced
986	# by "[[page_name]]". If it uses a special access URL beginning with "/w/", let it pass, as it
987	# probably cannot be replaced by "[[ ]]" markup
988	if [[ $URL == $WIKI_PATH ]] && [[ $URL != $WIKI_PATH/w/ ]]; then
989	STATUS="EI"
990	let EI_LINKS+=1
991	fi
992
993	# If it's not, check if this is a link to a domain that we have an interwiki prefix for (also make
994	# sure that it's not an archive.org link to a page from an interwiki domain)
995	if [ $STATUS == "??" ] && [[ $URL != web.archive.org ]]; then
996	for ((i = 0; i < ${#INTERWIKI_DOMAINS[@]}; ++i)); do
997	if [[ $URL == ${INTERWIKI_DOMAINS[$i]} ]] && [[ $URL != ${INTERWIKI_DOMAINS[$i]}/w/ ]]; then
998	STATUS="IW"
999	let IW_LINKS+=1
1000	INTERWIKI_INDEX=$i
1001	break
1002	fi
1003	done
1004	fi
1005
1006	# If we didn't match an interwiki domain, see if the status code is in our "OK" codes list
1007	if [ $STATUS == "??" ]; then
1008	for CODE in "${OK_CODES[@]}"; do
1009	if [[ $CODE == $CURL_CODE ]]; then
1010	STATUS="OK"
1011	let OK_LINKS+=1
1012
1013	# If this is a YouTube link, we have to look at the actual page source to know if the video
1014	# is good or not; override the link's info if it's actually NG
1015	if [[ $URL == www.youtube.com ]]; then
1016	PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $URL \| grep "\"simpleText\":\"Video unavailable\"")
1017	if [ ! -z "$PAGE_TEXT" ]; then
1018	STATUS="NG"
1019	CURL_RESULT=404
1020	let OK_LINKS-=1
1021	let NG_LINKS+=1
1022	fi
1023	fi
1024	break
1025	fi
1026	done
1027	fi
1028
1029	# If we didn't get a match with the "OK" codes, check it against the "RD" codes
1030	if [ $STATUS == "??" ]; then
1031	for CODE in "${RD_CODES[@]}"; do
1032	if [[ $CODE == $CURL_CODE ]]; then
1033	# Get URL header again in order to retrieve the URL we are being redirected to
1034	NEW_URL=$(curl -o /dev/null --silent --insecure --head --user-agent '$AGENT' --max-time $TIMEOUT --write-out '%{redirect_url}\n' $URL)
1035
1036	# Adjust the old and new URLs to both use HTTP for comparison purposes, so we can filter
1037	# those changes out if the user didn't ask for them
1038	URL_HTTP=$(echo $URL \| sed -E 's/^https:/http:/')
1039	NEW_URL_HTTP=$(echo $NEW_URL \| sed -E 's/^https:/http:/')
1040
1041	# Sometimes 'curl' fails to get the redirect_url due to time-out or bad web site config
1042	NEW_URL_LENGTH=$(echo \| awk -v input=$NEW_URL_HTTP '{print length(input)}')
1043	if [ $NEW_URL_LENGTH -lt $MIN_URL_LENGTH ]; then
1044	NEW_URL_HTTP="[new URL not retrieved]"
1045	fi
1046
1047	# Remove slash at end of new URL, if present, so we can filter out the redirects that
1048	# merely add an ending slash if the user didn't ask for them
1049	NEW_URL_NO_SLASH=$(echo $NEW_URL_HTTP \| sed -E 's:/$::')
1050
1051	# Detect if this is a youtu.be link simply being expanded by YouTube to the full
1052	# youtube.com address
1053	YOUTU_BE=0
1054	if [[ $URL_HTTP == http://youtu.be* ]] && [[ $NEW_URL_HTTP == http://www.youtube.com* ]]; then
1055	YOUTU_BE=1
1056	fi
1057
1058	# If the URLs match besides HTTP being upgraded to HTTPS, then the link is OK (unless user
1059	# wants those to be reported)
1060	if [ $SHOW_HTTPS -eq 0 ] && [ $URL_HTTP == $NEW_URL_HTTP ]; then
1061	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to show http->https upgrades, and I was redirected to '$NEW_URL'."
1062	STATUS="OK"
1063	let OK_LINKS+=1
1064	let SKIP_HTTPS_UP+=1
1065	# If the URLs match besides an added ending slash, then the link is OK (unless user wants
1066	# those to be reported)
1067	elif [ $SHOW_SLASH -eq 0 ] && [ $URL_HTTP == $NEW_URL_NO_SLASH ]; then
1068	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to show added trailing slashes, and I was redirected to '$NEW_URL'."
1069	STATUS="OK"
1070	let OK_LINKS+=1
1071	let SKIP_SLASH_ADD+=1
1072	elif [ $YOUTU_BE -eq 1 ]; then
1073	# We have to look at the actual page source to know if a YouTube video is good or not
1074	PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $NEW_URL \| grep "\"simpleText\":\"Video unavailable\"")
1075	if [ ! -z "$PAGE_TEXT" ]; then
1076	STATUS="NG"
1077	let NG_LINKS+=1
1078	else
1079	if [ $SHOW_YT_RD -eq 0 ]; then
1080	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to show redirects for youtu.be links, and I was redirected to '$NEW_URL'."
1081	STATUS="OK"
1082	let OK_LINKS+=1
1083	let SKIP_YOUTU_BE+=1
1084	else
1085	STATUS="RD"
1086	let RD_LINKS+=1
1087	fi
1088	fi
1089	else
1090	STATUS="RD"
1091	let RD_LINKS+=1
1092	fi
1093	break
1094	fi
1095	done
1096	fi
1097
1098	# If we didn't get a match with the "RD" codes, check it against the "NG" codes
1099	if [ $STATUS == "??" ]; then
1100	for CODE in "${NG_CODES[@]}"; do
1101	if [[ $CODE == $CURL_CODE ]]; then
1102	STATUS="NG"
1103	let NG_LINKS+=1
1104	break
1105	fi
1106	done
1107	fi
1108
1109	# If we didn't match a known status code, advise the reader
1110	if [ $STATUS == "??" ]; then
1111	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I encountered the unknown response code $CURL_CODE."
1112	let SKIP_UNK_CODE+=1
1113	continue
1114	fi
1115
1116	# Check problem links against exceptions list before proceeding
1117	FOUND_EXCEPT=0
1118	if [ $STATUS != "OK" ] && [ ! -z "$EXCEPT_URL" ]; then
1119	# The code we expect to find in the exceptions file is either the 'curl' result or "EI"/"IW"
1120	EXPECT_CODE="$CURL_RESULT"
1121	if [ $STATUS == "EI" ]; then
1122	EXPECT_CODE="EI"
1123	elif [ $STATUS == "IW" ]; then
1124	EXPECT_CODE="IW"
1125	fi
1126
1127	# Look for link in exceptions list and make sure the listed result code and wiki page also match
1128	for ((i = 0; i < ${#EXCEPT_ARRAY[@]}; ++i)); do
1129	{
1130	EXCEPT_LINE="${EXCEPT_ARRAY[$i]}"
1131
1132	# Undo any HTML-encoding from the wiki page; for now we just worry about the ampersand, as most
1133	# other HTML-encoded characters are not found in URLs
1134	EXCEPT_LINE=$(echo "$EXCEPT_LINE" \| sed 's/\&/\&/g')
1135
1136	# Check for URL match
1137	EXCEPT_URL="${EXCEPT_LINE#*,}"
1138	EXCEPT_URL="${EXCEPT_URL%,*}"
1139	if [ "$EXCEPT_URL" != "$URL" ]; then
1140	continue
1141	fi
1142
1143	# Check for page name match
1144	EXCEPT_PAGE="${EXCEPT_LINE##*,}"
1145	EXCEPT_PAGE="${EXCEPT_PAGE%% *}"
1146	if [ "$EXCEPT_PAGE" == "*" ] \|\| [ "$EXCEPT_PAGE" == "$LOCAL_PAGE_PATH" ]; then
1147	let EXCEPT_FOUND[$i]+=1
1148	valPrint trs "Found exception '$URL' on page '$LOCAL_PAGE_PATH'."
1149
1150	# Check for result code match
1151	EXCEPT_CODE=${EXCEPT_LINE%%,*}
1152	if [ "$EXCEPT_CODE" == "$EXPECT_CODE" ]; then
1153	FOUND_EXCEPT=1
1154	let EXCEPT_USED[$i]+=1
1155	valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because its expected result, '$EXPECT_CODE', is in the exceptions list."
1156
1157	if [ $STATUS == "EI" ]; then
1158	let SKIP_EXPECT_EI+=1
1159	elif [ $STATUS == "IW" ]; then
1160	let SKIP_EXPECT_IW+=1
1161	elif [ $STATUS == "RD" ]; then
1162	let SKIP_EXPECT_RD+=1
1163	else
1164	let SKIP_EXPECT_NG+=1
1165	fi
1166
1167	break
1168	fi
1169	fi
1170	} done
1171	fi
1172	if [ $FOUND_EXCEPT -eq 1 ]; then
1173	continue
1174	fi
1175
1176	# If appropriate, record this link to the log, with clickable URLs when possible
1177	if [ $STATUS != "OK" ] \|\| [ $RECORD_OK_LINKS -eq 1 ]; then
1178	# Prepare 'curl' result in parentheses to print after status code, unless this is an "IW" or "EI"
1179	# link, in which case showing the status code doesn't make sense. Adjust spacing after string to
1180	# ensure TXT and RTF reports have aligned columns of results.
1181	CURL_STR_H=" ($CURL_RESULT)"
1182	CURL_STR_T="$CURL_STR_H"
1183	CURL_STR_R="$CURL_STR_H "
1184	if [ $STATUS == "IW" ] \|\| [ $STATUS == "EI" ]; then
1185	CURL_STR_H=""
1186	CURL_STR_T=" "
1187	CURL_STR_R=" "
1188	fi
1189
1190	# Record link and its wiki page in TXT, RTF, and HTML markup
1191	valPrint t "${STATUS}${CURL_STR_T} $STR_TYPE $URL"
1192	valPrint t " linked from $FULL_PAGE_PATH"
1193	valPrint r "${STATUS}${CURL_STR_R}${RTF_TABS}$STR_TYPE {\field{\*\fldinst{HYPERLINK \"$URL\"}}{\fldrslt $URL}}"
1194	valPrint r " linked from {\field{\*\fldinst{HYPERLINK \"$FULL_PAGE_PATH\"}}{\fldrslt $LOCAL_PAGE_PATH}}"
1195	valPrint hn "<tr><td style=\"white-space:nowrap\">${STATUS}${CURL_STR_H}</td><td align=\"right\">$STR_TYPE</td><td><a href=\"$URL\" target=\"_blank\">$URL</a></td></tr>"
1196	valPrint hn "<tr><td colspan=\"2\" align=\"right\">linked from</td><td><a href=\"$FULL_PAGE_PATH\" target=\"_blank\">$LOCAL_PAGE_PATH</a></td></tr>"
1197
1198	# Place vertical space here since we won't be printing anything more about this link
1199	if [ $STATUS == "OK" ] && [ $SUGGEST_SNAPSHOTS_OK -eq 0 ]; then valPrint tr ""; valPrint hs ""; fi
1200
1201	# Record redirect URL if one was given by a 3xx response page
1202	if [ $STATUS == "RD" ]; then
1203	valPrint ts " Server suggests $NEW_URL"
1204	valPrint rs " Server suggests {\field{\*\fldinst{HYPERLINK \"$NEW_URL\"}}{\fldrslt $NEW_URL}}"
1205	valPrint hs "<tr><td colspan=\"2\" align=\"right\">Server suggests</td><td><a href=\"$NEW_URL\" target=\"_blank\">$NEW_URL</a></td></tr>"
1206	fi
1207
1208	# Notify reader if we can use an intrawiki link for this URL
1209	if [ $STATUS == "EI" ]; then
1210	INTRA_PAGE=${URL#:///}
1211	valPrint ts " Just use [[$INTRA_PAGE]]"
1212	valPrint rs " Just use [[$INTRA_PAGE]]"
1213	valPrint hs "<tr><td colspan=\"2\" align=\"right\">Just use</td><td>[[$INTRA_PAGE]]</td></tr>"
1214	fi
1215
1216	# Notify reader if we can use an interwiki prefix for this URL
1217	if [ $STATUS == "IW" ]; then
1218	INTER_PAGE=$(echo "$URL" \| sed 's/.*\///')
1219	valPrint ts " You can use [[${INTERWIKI_PREFIXES[$INTERWIKI_INDEX]}:$INTER_PAGE]]"
1220	valPrint rs " You can use [[${INTERWIKI_PREFIXES[$INTERWIKI_INDEX]}:$INTER_PAGE]]"
1221	valPrint hs "<tr><td colspan=\"2\" align=\"right\">You can use</td><td>[[${INTERWIKI_PREFIXES[$INTERWIKI_INDEX]}:$INTER_PAGE]]</td></tr>"
1222	fi
1223
1224	# Query Internet Archive for latest "OK" snapshot for "NG" page
1225	if [[ ( $STATUS == "NG" && $SUGGEST_SNAPSHOTS_NG -eq 1 ) \|\| ( $STATUS == "OK" && $SUGGEST_SNAPSHOTS_OK -eq 1 ) ]]; then
1226
1227	# We need to watch out for the rate limit or we'll get locked out; look at how much time has
1228	# elapsed and then wait the remainder between that and how long of a wait we think is needed
1229	# to avoid the dreaded "Too Many Requests" response. 5 seconds is just a guess.
1230	CUR_TIME=$(date +%s)
1231	WAIT_REMAINDER=$((5 - $CUR_TIME + $START_LINK))
1232	if [ $WAIT_REMAINDER -gt 0 ]; then
1233	valPrint t "Waiting $WAIT_REMAINDER second(s) to conform to Archive.org rate limit."
1234	sleep $WAIT_REMAINDER
1235	fi
1236
1237	# Issue query to the API
1238	ARCHIVE_QUERY=$(curl --silent --max-time $TIMEOUT "$ARCHIVE_API?url=$URL&$ARCHIVE_OK_CODES")
1239
1240	# Notify reader if we hit the rate limit and just keep going
1241	if [[ "$ARCHIVE_QUERY" == "Too Many Requests" ]]; then
1242	valPrint t " IA has rate-limited us!"
1243	valPrint r " IA has rate-limited us!"
1244	valPrint hs "<tr><td colspan=\"2\" align=\"right\">IA suggests</td><td>(hit the API rate limit!)</td></tr>"
1245	# If a "closest" snapshot was received, inform reader
1246	elif [[ "$ARCHIVE_QUERY" == \"closest\": ]]; then
1247	# In case the URL has a shebang in it (like mega.nz links do), escape the '!' to break it
1248	ARCHIVE_QUERY=$(echo "$ARCHIVE_QUERY" \| sed 's/#!/#\\!/')
1249
1250	# ...isolate "url" property in the response that follows the "closest" tag
1251	SNAPSHOT_URL=${ARCHIVE_QUERY##*\"closest\":} # everything after '"closest":'
1252	SNAPSHOT_URL=${SNAPSHOT_URL#*\"url\": \"} # everything after '"url": "'
1253	SNAPSHOT_URL=${SNAPSHOT_URL%%\"*} # everything before '"'
1254
1255	# Remove the port 80 part that IA often adds to the URL, as it's superfluous
1256	SNAPSHOT_URL=$(echo $SNAPSHOT_URL \| sed 's/:80//')
1257
1258	# Inform the reader of the snapshot URL
1259	valPrint ts " IA suggests $SNAPSHOT_URL"
1260	valPrint rs " IA suggests {\field{\*\fldinst{HYPERLINK \"$SNAPSHOT_URL\"}}{\fldrslt $SNAPSHOT_URL}}"
1261	valPrint hs "<tr><td colspan=\"2\" align=\"right\">IA suggests</td><td><a href=\"$SNAPSHOT_URL\" target=\"_blank\">$SNAPSHOT_URL</a></td></tr>"
1262	else # Otherwise give a generic Wayback Machine link for this URL, which might work
1263	valPrint ts " Try browsing $ARCHIVE_GENERIC/$URL"
1264	valPrint rs " Try browsing {\field{\*\fldinst{HYPERLINK \"$ARCHIVE_GENERIC/$URL\"}}{\fldrslt $ARCHIVE_GENERIC/$URL}}"
1265	valPrint hs "<tr><td colspan=\"2\" align=\"right\">Try browsing</td><td><a href=\"$ARCHIVE_GENERIC/$URL\" target=\"_blank\">$ARCHIVE_GENERIC/$URL</a></td></tr>"
1266	fi
1267	fi
1268	fi
1269
1270	# If this is a page and it seems to be there, take screenshot of it using headless Google Chrome
1271	if [ $IS_FILE -eq 0 ] && [ $TAKE_PAGE_SHOT -eq 1 ] && [ $STATUS == "OK" ]; then
1272	# Sanitize URL by removing "http(s)://" and converting any other colons or slashes to underscores
1273	SHOT_NAME=$(echo "$URL" \| sed 's/https*\:\/\///' \| sed 'y/:\//__/')
1274	SHOT_FILE="$SHOT_PATH/$SHOT_NAME.png"
1275
1276	# Don't take screenshot if we already encountered this page and screenshotted it
1277	if [ ! -f "$SHOT_FILE" ]; then
1278	"$CHROME_PATH" --headless --disable-gpu --screenshot --window-size=1500,900 $URL > /dev/null 2>&1
1279	if [ -f "$WORKING_DIR/$CHROME_SCREENSHOT" ]; then
1280	mv -n "$WORKING_DIR/$CHROME_SCREENSHOT" "$SHOT_FILE"
1281	else
1282	valPrint trhs "Screenshot of URL $URL seems to have failed!"
1283	fi
1284	else
1285	valPrint trhs "Skipping screenshot of URL '$URL' because file '$SHOT_FILE' already exists."
1286	fi
1287	fi
1288	done
1289	FINISHED_LIST="yes"
1290	wrapupAndExit

Note: See TracBrowser for help on using the repository browser.

Download in other formats: