Merge branch 'master' of ssh://git.thebackupbox.net/uritools

14 files changed, 328 insertions, 50 deletions

diff --git a/Makefile b/Makefile
index fdbd973..090e2c0 100644
--- a/Makefile
+++ b/Makefile
@@ -24,6 +24,10 @@ clean:
 	rm -f *.o
 
 install: all
+	install -t $(PREFIX)/bin sqesc
+	install -t $(PREFIX)/bin ssh_hack
+	install -t $(PREFIX)/bin xwindowURI
+	install -t $(PREFIX)/bin uriqueryfilter
 	install -t $(PREFIX)/bin urimatch
 	install -t $(PREFIX)/bin uricut
 	install -t $(PREFIX)/bin urijoin
diff --git a/README b/README
index 1ea4b28..63707b6 100644
--- a/README
+++ b/README
@@ -1,16 +1,9 @@
-This is a handy little program to help you handle URIs that any program may want a handler for.
-Just point those programs looking for a handler at uristart.
+This repo has grown a lot of tools that I use on my own system.
 
-Then write a matching pattern, and a handler pipeline.
+There's a uri.h that can be included in other C programs
 
-The matching patterns are a series of pairs of URI-piece and a string.
-
-Each piece has a single letter you use to select the piece.
-
-here's an example that uses all of the parts:
-s://u:k@d:P/p?q#f
-
-where each piece is:
+Most of the programs will take single letters to refer to parts of a URI.
+each piece is:
 s: scheme
 u: username
 k: password (k for key)
@@ -19,3 +12,33 @@ P: port
 p: path
 q: query_string
 f: fragment_id
+
+## uricut
+
+used for cutting portions of a URI out from the rest of it.
+
+to cut the domain out of all the input URIs and show one per line.
+```
+urilist | uricut -d
+```
+
+## urimatch
+
+used for matching URIs
+multiple arguments will print if any of the rules match.
+so if you want to print only if all of the rules match, pipe urimatch into another urimatch
+
+to print all gemini or gopher URIs:
+```
+urilist | urimatch s gemini s gopher
+```
+
+to print all the gemini URIs that are for the domain gemini.thebackupbox.net
+```
+urilist | urimatch s gemini | urimatch d gemini.thebackupbox.net
+```
+
+to print all the non-http links using reverse match and globbing
+```
+urilist | urimatch rs 'http*'
+```
diff --git a/choose b/choose
index be4602a..1ec7dd7 100755
--- a/choose
+++ b/choose
@@ -7,7 +7,9 @@
 #shift
 #sed 's|,|\\,|g;s|:|\\:|g' | tr '\n' ',' | rev | cut -b2- | rev | tr '\n' '\0' | xargs -0 xmessage -nearmouse "$@" "$message" -print -buttons
 if [ $DISPLAY ];then
-  dmenu -l 10 -p "$1"
+#usage: dmenu [-bfiv] [-l lines] [-p prompt] [-fn font] [-m monitor]
+#             [-nb color] [-nf color] [-sb color] [-sf color] [-w windowid]
+  dmenu -l 10 -p "$1" -nb '#000000' -nf '#00aa00' -sb '#006600' -sf '#00ff00'
 else
-  printf "%s" "$1" ; ledit head -n1
+  printf "%s" "$1"
 fi
diff --git a/copy_start_nevermind.sh b/copy_start_nevermind.sh
index 3f7f7af..66bf922 100755
--- a/copy_start_nevermind.sh
+++ b/copy_start_nevermind.sh
@@ -1,16 +1,35 @@
 #!/bin/bash
 [ "$1" ] || exit 1
 uri="$1"
+
+queryfiltereduri="$(uriqueryfilter "$uri")"
+if [ "$uri" != "$queryfiltereduri" ];then
+  trackeruri="$uri"
+  uri="$queryfiltereduri"
+fi
+
+### this clipping should be part of choose probably
 if [ "${uri}" != "${uri:0:64}" ];then
   prompt="${uri:0:64}..."
+else
+  prompt="${uri}"
 fi
+
 scheme="$(uricut -s <<< "$uri")"
-selection="$(printf "start with: %s\ncopy\nshorten\nverbose\nforget\nQR\n%s\n" "$(urigetline "$uri")" "$(printf "%s\n" "$uri" | uricut | grep -v '^whole')" \
+selection="$(printf "start with: %s\nremote_csn\ncopy\nshorten\nunshorten\ntitle\nverbose\nforget\nQR\n%s\n" "$(urigetline "$uri")" "$(printf "%s\n" "$uri" | uricut | grep -v '^whole')" \
              | choose "${prompt}")"
 #            | choose "$(printf "uri: '%s'\nwill be ran with: %s\n" "$uri" "$(urigetline "$uri")" )" -default start )"
 echo "$selection"
 case "$selection" in
 
+remote_csn)
+  remote_host="$(choose "remote host to run URI on> " < ~/.config/urirhosts.conf)"
+  ### TODO: include in the list our own display either to use -X or use our IP and display number?
+  display="$(ssh "${remote_host}" myXdisplays | choose "which display?")"
+  uristart "ssh://${remote_host}/copy_start_nevermind.sh%20${uri}?DISPLAY=$display"
+  exit 0
+  ;;
+
 verbose)
   x-terminal-emulator -e less -f <(printf "%s" "$uri")
   exit 0
@@ -21,6 +40,11 @@ copy)
   exit 0
   ;;
 
+title)
+  copy_start_nevermind.sh "$(urititle "$uri")"
+  exit 0
+  ;;
+
 forget)
   cp ~/.cache/uristart.log ~/.cache/uristart.log.tmp
   cat ~/.cache/uristart.log.tmp | grep -vFw "${uri}" > ~/.cache/uristart.log
@@ -44,11 +68,18 @@ shorten)
   exit 0
   ;;
 
+unshorten)
+  copy_start_nevermind.sh "$(unshorten.sh "$uri")"
+  exit 0
+  ;;
+
 QR)
   x-terminal-emulator -hold -e qr "$uri"
   ;;
 
 *)
-  exec printf "%s" "$selection" | cut '-d ' -f2- | xclip -i
+  if [ "$selection" ];then
+    exec printf "%s" "$selection" | cut '-d ' -f2- | xclip -i
+  fi
   ;;
 esac
diff --git a/search b/search
index 1c2e67c..7c8c77a 100755
--- a/search
+++ b/search
@@ -1,2 +1,6 @@
 #!/bin/sh
-uristart "search:?q=$(echo | dmenu -p 'search>' | tr '\n' '\0' | xargs -0 uriescape | sed 's/&/%26/g;s/#/%23/g')"
+q="$(printf "%s\n%s\n" "$(xclip -o | head -n1)" "$(xclip -o -selection clipboard | head -n1)" | choose 'search>' | tr '\n' '\0' | xargs -0 uriescape | sed 's/&/%26/g;s/#/%23/g')"
+if [ ! "$q" ];then
+  exit 1
+fi
+exec uristart "search:?q=$q"
diff --git a/sqesc.c b/sqesc.c
new file mode 100644
index 0000000..6d7ba23
--- /dev/null
+++ b/sqesc.c
@@ -0,0 +1,18 @@
+#include <stdio.h>
+#include <string.h>
+
+int print_escaped(char *s) {
+  char *p=0;
+  for(p=strchr(s,'\'');p && s && *p && *s;p=strchr(s,'\'')) {
+    *p=0;//null out this single-quote
+    printf("%s",s);
+    printf("'\\''");
+    s=p+1;
+  }
+  printf("%s",s);
+  return 0;
+}
+
+int main(int argc,char *argv[]) {
+  return print_escaped(argv[1]);
+}
diff --git a/ssh_hack b/ssh_hack
new file mode 100755
index 0000000..e882cd2
--- /dev/null
+++ b/ssh_hack
@@ -0,0 +1,26 @@
+#!/usr/bin/env bash
+### we need to export each of the query string variables
+### then send them over ssh
+q="$(printf "%s\n" "$1" | uricut -q)"
+p="$(printf "%s\n" "$1" | uricut -p)"
+d="$(printf "%s\n" "$1" | uricut -d)"
+P="$(printf "%s\n" "$1" | uricut -P)"
+u="$(printf "%s\n" "$1" | uricut -u)"
+
+eval "$(printf '%s\n' "$q" | tr '&' ' ' | grep '[^ ]*=' | sed 's/^/export /g')"
+if [ "$p" ];then
+  p="$(uriunescape "$p")"
+fi
+if [ "$P" ]; then
+  HACK_PORT="-p $P"
+fi
+if [ "$u" ];then
+  HACK_USER="$u@"
+fi
+
+tmpfile=$(mktemp);
+printf '%s\n' "$q" | tr '&' '\n' | cut -d= -f1 | sed 's/^/SendEnv /g' > "$tmpfile"
+
+ssh -F "$tmpfile" -t ${HACK_PORT} ${HACK_USER}${d} ${p:1}
+#ssh -F <(printf '%s\n' "$q" | tr '&' '\n' | cut -d= -f1 | sed 's/^/SendEnv /g') -t ${HACK_PORT} ${HACK_USER}${d} ${p:1}
+#ssh -F <(printf '%s\n' "$q" | tr '&' '\n' | cut -d= -f1 | sed 's/^/SendEnv /g') -t ${HACK_PORT} ${HACK_USER}${d} ${p:1}
diff --git a/urilaunch b/urilaunch
index 88b2a40..fc15323 100755
--- a/urilaunch
+++ b/urilaunch
@@ -3,10 +3,12 @@
 #you'll probably want to change my gopher-get line to be something useful to you.
 exec copy_start_nevermind.sh \
   "$(cat \
-      <(printf "%s\n" $(xclip -o | head -n1)) \
+      <(printf "%s\n" "$(xclip -o | head -n1)") \
+      <(printf "%s\n" "$(xclip -o -selection clipboard | head -n1)") \
+      <(xdotool getwindowfocus | xargs xwindowURI) \
       <(printf "\n") \
       <(gopher-get gopher://batou/0/links.txt) \
       <(cat ~/.cache/uristart.log | sort | uniq) \
-    | dmenu -l 10 -p urilaunch\> \
+    | choose 'urilaunch>' \
     | cut '-d ' -f1
   )"
diff --git a/urimatch.c b/urimatch.c
index a90fb68..91177f2 100644
--- a/urimatch.c
+++ b/urimatch.c
@@ -6,29 +6,43 @@
 
 #define LINE_LENGTH 1024
 
-int match(char negate,char *part,char *arg) {
- if(negate) {
-  if(part == 0) return 1;//we found that the part isn't here!
- } else {
-  if(part) {
-   if(!fnmatch(arg,part,FNM_NOESCAPE)) return 1;
-  }
+#define MATCH_UNEXIST 0
+#define MATCH_PATTERN 1
+#define MATCH_REVERSE -1
+
+// return 1 if the match and rule should have the main function print this URL.
+int match(char rule,char *part,char *arg) {
+ switch(rule) {
+  case MATCH_UNEXIST:
+   if(part == 0) return 1;
+   break;
+  case MATCH_REVERSE:
+   if(part) {
+     if(!fnmatch(arg,part,FNM_NOESCAPE)) return 0;
+     else return 1;
+   }
+   break;
+  case MATCH_PATTERN:
+   if(part) if(!fnmatch(arg,part,FNM_NOESCAPE)) return 1;
+   break;
+  default:
+   fprintf(stderr,"oh god. what the hell happened to get here?\n");
+   break;
  }
  return 0;
 }
 
 int main(int argc,char *argv[]) {
- int i;
+ int i,j;;
  int ret=1;
  struct uri u;
- char negate=0;
+ char rule=MATCH_PATTERN;
  char *line=malloc(LINE_LENGTH);
  char copy[LINE_LENGTH];
  if(argc < 2) {
-  printf("usage: urimatch [-][n][s|u|k|d|D|P|p|q|f] [string]\n");
+  printf("usage: urimatch [-][n][r][s|u|k|d|P|p|q|f] [string]\n");
   printf("scheme://username:password@domain:port/path?query_string#fragment_id\n");
   printf("s://u:k@d:P/p?q#f\n");
-  printf("The D flag is special. it matches its argument against the last bytes of the input url's domain.\n");
   printf("This allows matching of subdomains, like `echo epoch.ano | urimatch -D ano` would match.\n");
   printf("the 'n' flag can be put before any of the other flags to check for a missing.\n");
   return 1;
@@ -40,30 +54,31 @@ int main(int argc,char *argv[]) {
   memset(&u,0,sizeof(u));
   urifromline(&u,line);
   //use the character in argv[1] to match stdin against argv[2]. if match print whole line.
-  for(i=1;i<argc;i+=2) {
-   if(negate) {i--;}//we didn't really need to go that far.
-   negate=0;
-   if(argv[i][0] == '-') argv[i]++;
-   if(argv[i][0] == 'n') {argv[i]++; negate=1; }//heh.
+  for(i=1;i<argc;i+=2) {//for each line, we're reparsing the arguments...
+   rule=MATCH_PATTERN;//default
+   j=0;
    switch(argv[i][0]) {
-    case 's': if(match(negate,u.scheme,argv[i+1]))       { printf("%s\n",copy); ret=0;} break;
-    case 'u': if(match(negate,u.username,argv[i+1]))     { printf("%s\n",copy); ret=0;} break;
-    case 'k': if(match(negate,u.password,argv[i+1]))     { printf("%s\n",copy); ret=0;} break;
-    case 'd': if(match(negate,u.domain,argv[i+1]))       { printf("%s\n",copy); ret=0;} break;
-    case 'P': if(match(negate,u.port,argv[i+1]))         { printf("%s\n",copy); ret=0;} break;
-    case 'p': if(match(negate,u.path,argv[i+1]))         { printf("%s\n",copy); ret=0;} break;
-    case 'q': if(match(negate,u.query_string,argv[i+1])) { printf("%s\n",copy); ret=0;} break;
-    case 'f': if(match(negate,u.fragment_id,argv[i+1]))  { printf("%s\n",copy); ret=0;} break;
-    case 'D': //not sure how to look for a missing one of these. it'd be like d.
-     if(u.domain && argv[i+1] && strlen(u.domain) >= strlen(argv[i+1]) && !strncmp(u.domain+strlen(u.domain)-strlen(argv[i+1]),argv[i+1],strlen(argv[i+1]))) {
-      printf("%s\n",copy);
-      ret=0;
-     }
-     break;
+     case '-': j=1; rule=MATCH_PATTERN; break;//so we can't be modifying these! how the fuck?
+     case 'n': j=1; rule=MATCH_UNEXIST; break;
+     case 'r': j=1; rule=MATCH_REVERSE; break;
+     default: break;//handled by the next switch
+   }
+   switch(argv[i][j]) {
+    case 's': if(match(rule,u.scheme,argv[i+1]))       { printf("%s\n",copy); ret=0;} break;
+    case 'u': if(match(rule,u.username,argv[i+1]))     { printf("%s\n",copy); ret=0;} break;
+    case 'k': if(match(rule,u.password,argv[i+1]))     { printf("%s\n",copy); ret=0;} break;
+    case 'd': if(match(rule,u.domain,argv[i+1]))       { printf("%s\n",copy); ret=0;} break;
+    case 'P': if(match(rule,u.port,argv[i+1]))         { printf("%s\n",copy); ret=0;} break;
+    case 'p': if(match(rule,u.path,argv[i+1]))         { printf("%s\n",copy); ret=0;} break;
+    case 'q': if(match(rule,u.query_string,argv[i+1])) { printf("%s\n",copy); ret=0;} break;
+    case 'f': if(match(rule,u.fragment_id,argv[i+1]))  { printf("%s\n",copy); ret=0;} break;
+    // TODO: after a while, get rid of 'D' completely.
+    case 'D': fprintf(stderr,"the 'D' flag has been removed.\ninstead of: D 'host'\nuse: d '*host'\n");
     default:
      printf("unknown url part letter! '%c'\n",argv[i][0]);
      return ret;
    }
+   if(rule == MATCH_UNEXIST) {i--;}//UNEXIST doesn't consume the argument after it. so, go backwards 1. we're about to go ahead 2.
   }
  }
  return ret;
diff --git a/uriqueryfilter b/uriqueryfilter
new file mode 100755
index 0000000..72dca47
--- /dev/null
+++ b/uriqueryfilter
@@ -0,0 +1,101 @@
+#!/usr/bin/env bash
+## list grabbed from:
+## https://github.com/newhouse/url-tracking-stripper/blob/master/assets/js/trackers.js
+## referenced from:
+## https://github.com/rknightuk/TrackerZapper/blob/3fe21dc2a59240dc2883569f8d435415010610d2/TrackerZapper/AppDelegate.swift#L162
+## more:
+## https://maxchadwick.xyz/tracking-query-params-registry/
+
+## usages:
+##   ./paramfilter URL1 URL2 URL3 ...
+##   ./paramfilter < list.uris
+## if you use arguments it won't try to read from stdin.
+## so if you want to use paramfilter as a filter don't use arguments.
+
+if [ "$1" ];then
+  while [ "$1" ];do
+    printf "%s\n" "$1"
+    shift
+  done
+else
+  cat #useless cat?
+fi | while read line;do
+  q="$(uricut -q "$line")"
+  if [ "$q" ];then
+    newq="$(printf "%s\n" "$q" | tr '&' '\n' | grep -vf <(grep -A 1000 "^__LIST__$" "$0" | tail -n+2 | sed 's/^/^/g;s/$/=/g') | tr '\n' '&' | sed 's/&$//')"
+    cat <(uricut "$line") <(printf "query_string: %s\n" "$newq") | urijoin 2>/dev/null
+  else
+    printf "%s\n" "$line"
+  fi
+done
+
+exit 0
+#this was generated with:
+# wget -qO- https://github.com/mpchadwick/tracking-query-params-registry/raw/master/_data/params.csv | tail -n+2 | cut -d, -f1 >> uriqueryfilter
+# echo ref_src >> uriqueryfilter
+__LIST__
+fbclid
+gclid
+gclsrc
+utm_content
+utm_term
+utm_campaign
+utm_medium
+utm_source
+utm_id
+_ga
+mc_cid
+mc_eid
+_bta_tid
+_bta_c
+trk_contact
+trk_msg
+trk_module
+trk_sid
+gdfms
+gdftrk
+gdffi
+_ke
+redirect_log_mongo_id
+redirect_mongo_id
+sb_referer_host
+mkwid
+pcrid
+ef_id
+s_kwcid
+msclkid
+dm_i
+epik
+pk_campaign
+pk_kwd
+pk_keyword
+piwik_campaign
+piwik_kwd
+piwik_keyword
+mtm_campaign
+mtm_keyword
+mtm_source
+mtm_medium
+mtm_content
+mtm_cid
+mtm_group
+mtm_placement
+matomo_campaign
+matomo_keyword
+matomo_source
+matomo_medium
+matomo_content
+matomo_cid
+matomo_group
+matomo_placement
+hsa_cam
+hsa_grp
+hsa_mt
+hsa_src
+hsa_ad
+hsa_acc
+hsa_net
+hsa_kw
+hsa_tgt
+hsa_ver
+ref_src
diff --git a/uristart b/uristart
index a5c3e63..2b0d7f2 100755
--- a/uristart
+++ b/uristart
@@ -10,6 +10,8 @@ ASKUSER="choose"
 
 check=dont
 
+eval "$(printf "%s\n" "$1" | uricut | sed 's/^/export URI/;s/'\''/'\''\\'\'''\''/;s/: /='\''/g;s/$/'\''/')"
+
 uristart_config=~/.config/uristart.conf
 
 usage() {
@@ -43,7 +45,7 @@ if [ "$verbose" ];then
   echo "uristart: other:" "$@" >&2
 fi
 
-if [ ! "$@" ];then
+if [ ! "$*" ];then
   usage >&2 #to stderr because it is an error to not have any extra args to process
   exit 1
 fi
diff --git a/uriunescape.c b/uriunescape.c
index 23800a5..2261da9 100644
--- a/uriunescape.c
+++ b/uriunescape.c
@@ -11,7 +11,7 @@ int main(int argc,char *argv[]) {
   char buf2[BUFSIZE];
   char *ab;
   int blen;
-  int hack;
+  int hack=0;
   if(argc > 1) {
     for(argv++,argc--;argc;argc--,argv++) {
       ab=malloc(strlen(*argv));
@@ -40,6 +40,7 @@ int main(int argc,char *argv[]) {
         }
       }
     }
+    if(blen < 0) perror("read");
   }
   return 0;
 }
diff --git a/urnresolve b/urnresolve
index fc777a4..d59558f 100755
--- a/urnresolve
+++ b/urnresolve
@@ -25,6 +25,16 @@ if [ "$nid" = "ietf" ];then
   fi
 fi
 
+if [ "$nid" = "phrack" ];then
+  if [ "$nss" ];then
+    if [ "$nss2" ];then
+      echo "gopher://thebackupbox.net:70/0/phrack/issues/${nss}/${nss2}.txt"
+    else
+      echo "gopher://thebackupbox.net:70/1/phrack/issues/${nss}/"
+    fi
+  fi
+fi
+
 if [ "$nid" = "xkcd" ];then
   echo "https://xkcd.com/${nss}/"
 fi
@@ -40,3 +50,7 @@ if [ "$nid" = "fcc" ];then
     grep "^$nss2\s" $PREFIX/share/urn/fcc:frs | cut -f2
   fi
 fi
+
+if [ "$nid" = "btih" ];then
+  btih2magnet "$nss"
+fi
diff --git a/xwindowURI b/xwindowURI
new file mode 100755
index 0000000..6e9235a
--- /dev/null
+++ b/xwindowURI
@@ -0,0 +1,35 @@
+#!/bin/sh
+if [ ! "$1" ];then
+  id="$(xwinID)"
+else
+  id="$1"
+fi
+
+class=$(xclass -id "$id")
+#echo "$class"
+
+case "$class" in
+mupdf)
+  basename="$(xwintitle "$id" | rev | cut '-d ' -f5- | rev)"
+  echo "b: $basename"
+  page="$(xwintitle "$id" | tr ' ' '\n' | grep '[0-9][0-9]*/[0-9][0-9]*' | cut -d/ -f1)"
+  echo "$page"
+  for PID in $(pgrep mupdf);do
+    argv1="$(cat /proc/$PID/cmdline | tr '\0' '\n' | sed -n 2p)"
+    cd "$(readlink /proc/$PID/cwd)"
+    abspath="$(normalpath "$argv1")"
+    echo "a: $abspath"
+    echo "b: $basename"
+    if [ "$(basename "$abspath")" = "$basename" ];then
+      printf "file://%s#page=%s\n" "$(uriescape "$abspath")" "$page"
+      exit 0
+    fi
+  done
+  ;;
+Navigator)
+  xwintitle "$id" | rev | cut '-d ' -f4 | rev
+  ;;
+*)
+  echo "class: $class"
+  ;;
+esac