pbh actually looks good including tables

tables without thead dont render
cleanup table displays
2022-07-04 11:45:55 -06:00 · 2022-07-04 11:27:44 -06:00 · 2022-07-04 10:25:18 -06:00 · 2022-07-03 23:54:43 -06:00 · 2022-07-03 23:31:59 -06:00 · 2022-07-03 21:18:20 -06:00
3 changed files with 137 additions and 58 deletions
--- a/README.md
+++ b/README.md
@@ -0,0 +1,10 @@
+# dnd beyond scrape
+
+* run `SEED=123 bash scrape.sh`
+* see cached files, cookie, output in $TMPDIR/scrape.$SEED
+* may have to replace `Cookie: ` in scrape.sh if expired
+
+## Todo
+
+* fix tables to pre-formatted
+* download and localize images (webp!)
--- a/scrape.sh
+++ b/scrape.sh
@@ -3,7 +3,7 @@
 main() {
   ensure

-   local output="${1:-"$TMPDIR/result"}"
+   local output="$TMPDIR/result"
   mkdir -p "$output"
   log "$output"

@@ -55,7 +55,7 @@ ensure() {
         return 1
      fi
   done
-   export TMPDIR=/tmp/scrape.${SEED:-${RANDOM:-$(date +%s)}}
+   export TMPDIR="${TMPDIR:-"/tmp/scrape.${SEED:-${RANDOM:-$(date +%s)}}"}"
   mkdir -p $TMPDIR
 }

@@ -71,7 +71,8 @@ scrape_books() {
      | sed 's/^href="//' \
      | sed 's/^\///' \
      | sed 's/^/\//' \
-      | sed 's/^/https:\/\/www.dndbeyond.com/'
+      | sed 's/^/https:\/\/www.dndbeyond.com/' \
+      | grep "${SCRAPE_BOOKS_PATTERN:-.*}"
 }

 scrape_book() {
@@ -94,10 +95,23 @@ scrape_book() {
      fi
      scrape_book_chapter "$chapter" > "$f"
      log scraping book title from $f
-      local title="$(grep '^\#' "$f" | head -n 1 | sed 's/^\#*//' | sed 's/^[ ]*//' | sed 's/[ ]*$//')"
+      local title="$(
+         grep '^\#' "$f" \
+         | head -n 1 \
+         | sed 's/^\#*//' \
+         | sed 's/^[ ]*//' \
+         | sed 's/[ ]*$//' \
+         | sed 's/[ ][ ]*/ /g' \
+         | sed 's/[^a-zA-Z0-9]/_/g' \
+         | sed 's/mdash.*//' \
+         | sed 's/Disclaimer.*/Disclaimer/g' \
+      )"
      if [ -z "$title" ]; then
-         log "WARNING: NO TITLE FOUND IN $f ($chapter): $(head -n 3 $f)"
-         return 1
+         title="${chapter%/}"
+         title="${title##*/}"
+         title="${title%%"?"*}"
+         log "WARNING: NO TITLE FOUND IN $f ($chapter): $(head -n 3 $f); using $title from $chapter (enter to continue)"
+         read
      fi
      mv "$f" "$output"/"$title"
   done
@@ -181,6 +195,7 @@ _scrape_dndbeyond_cache_f() {
 _scrape_dndbeyond() {
   log foo:_scrape_dndbeyond $@
   rate_limit_1s scrape_dndbeyond
+   touch "$TMPDIR/cookies.txt"
   curl -L -sS "$1" \
      -H 'authority: www.dndbeyond.com' \
      -H 'cache-control: max-age=0' \
@@ -193,28 +208,12 @@ _scrape_dndbeyond() {
      -H 'sec-fetch-user: ?1' \
      -H 'sec-fetch-dest: document' \
      -H 'accept-language: en-US,en;q=0.9' \
-      -H 'Cookie: Preferences=undefined; Preferences=undefined; _pxhd=qvJb-g60h1HSymUFEiAmb-Noh35mNbUmcFgIE0qEiisCj/BqXrgrrsRUlODrjAaSGknpuHBn0S2b4rGTBFvaIA==:5/4beZX4ojNLKbAeZH9rQFcGIQ/ijS7lK29wvhLkgB4O0L2GFpRuU5FOpRztEiKg9EoUZr2cPB2Z8w3DKVkaKJPlwcIt23DM5PgQLZEPoA8=; ResponsiveSwitch.DesktopMode=1; LoginState=0ae9ef26-a1ae-4cb2-8af7-f18c21b99811; G_ENABLED_IDPS=google; CobaltSession=eyJhbGciOiJkaXIiLCJlbmMiOiJBMTI4Q0JDLUhTMjU2In0..FawE7bpRlluZsWq7Iny1xQ.nBX_qdZq5varuXNGkl70QeJ03JIRT3sxxQi6iCzazt-SjlK1hEuINX0sK85gJKVT.8wlUwv…sername=squeaky2x3; UserInfo={"UserId":109926924,"UserJoinDate":"2021-03-25","UserSessionId":"5ce909a9-34aa-4b15-9b6d-00ee48e13e05"}; ddbSiteBanner:bb279d28-f265-49d5-ad41-6d716915bbdf=true; AWSELB=17A593B6CA59C3C4856B812F84CD401A582EF083E14A2E133FB99235E1D3CC671549BC8DF3DD0A01DD4815C023F988F55E0E95275F44EAFDFB8393A8DEFA33DD0B90284D; AWSELBCORS=17A593B6CA59C3C4856B812F84CD401A582EF083E14A2E133FB99235E1D3CC671549BC8DF3DD0A01DD4815C023F988F55E0E95275F44EAFDFB8393A8DEFA33DD0B90284D; WarningNotification.Lock=1' \
+      -H 'Cookie: Preferences=undefined; Preferences=undefined; _pxhd=dGa4DRIksfo8m7Siqu0z-stTfu9R0jC2I53wbDc/nEiS/YpOxaQaWLECaBu4-wCfqyCDmn2c/PpjW1ESR3pWow==:RH/b5abeFbAM04R3V7bm14eVCFGzYpPFEt6-0X4z2CCnEnFxcFIQkGBwdews8JTWKuNsgYx758Xp3RcXupSxnPMBLfFLJ8MSNyzyaOozezE=; Geo={%22region%22:%22UT%22%2C%22country%22:%22US%22%2C%22continent%22:%22NA%22}; ResponsiveSwitch.DesktopMode=1; RequestVerificationToken=2cdddc90-fbbb-45a4-929a-1fff6879a852; sublevel=ANON; Preferences=undefined; marketplace_filter_show_owned=false; G_ENABLED_IDPS=google; CobaltSession=eyJhbGciOiJkaXIiLCJlbmMiOiJBMTI4Q0JDLUhTMjU2In0..YZH4-9MvXBhf0v5KdkH8vA.J5jwdmKCBGfAjmlGY1UOr6UD0am0hS0qiyY-xBx-j3CEM8CrhlzyxRDo04YK3nc3.Z5M5dZnstMa9OIkrEN29HQ; User.ID=109926924; User.Username=squeaky2x3; Preferences.Language=1; UserInfo={"UserId":109926924,"UserJoinDate":"2021-03-25","UserSessionId":"71f1f0eb-3523-4f94-96d9-22d3511ea6b6"}; Preferences.TimeZoneID=1; LoginState=8ab45ac1-5777-4613-a3f5-7439058fec58; AWSELB=17A593B6CA59C3C4856B812F84CD401A582EF083AB9FE2016C3192A7805F9642538006F9B284A9F525C1AA0DF220CB30AEE9DCF61CB6C98E40F1ADC8CB1E6C93F2D5E4FC; AWSELBCORS=17A593B6CA59C3C4856B812F84CD401A582EF083AB9FE2016C3192A7805F9642538006F9B284A9F525C1AA0DF220CB30AEE9DCF61CB6C98E40F1ADC8CB1E6C93F2D5E4FC; cobalt-token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.ygIntjw5eL4hlACI7T0VpdWKUghzaJZtrGsVllsgEtk; WarningNotification.Lock=1' \
      -H 'dnt: 1' \
+      -b "$TMPDIR/cookies.txt" \
+      -c "$TMPDIR/cookies.txt" \
      --compressed
   return
-   curl -L -sS \
-      "$1" \
-      --compressed \
-      -H 'Cookie: Preferences=undefined; Preferences=undefined; _pxhd=NHdkEjOXnAEHojER86i0Egmnm85GLvifiQxURPISwVtzWLNwuqQbIk1Y/2MjheQqNP4CsBXhuc4UG5-qcmLjYw==:hGx3TOEjeVaJPMfrXXnmyG7-btpQhZutwvpVBKwEbNXTAkEQvgmIs/faKB6s8537Nivj6epH9HpOWYURQeazeMwSMReW6EMazW09Rci-PJI=; ResponsiveSwitch.DesktopMode=1; ddbSiteBanner:00523ca3-81a4-4d2a-8f86-9e40273af2e2=true; LoginState=2c011278-8a3d-4cf7-a141-221f42694333; G_ENABLED_IDPS=google; CobaltSession=eyJhbGciOiJkaXIiLCJlbmMiOiJBMTI4Q0JDLUhTMjU2In0..vh9-_2G_aEjMvJ2PFAAE4w.eheZuOJLp-3NyiAJZzKqaSc91lH0ganCRsDZPYMGIG3265AppaHh0uMQLu_-P6WX.bHRi_gX8PKtQGZwVmNeb1Q; Preferences.Language=1; Preferences.TimeZoneID=1; ddbSiteBanner:d8879cae-c68e-4411-9ea0-a49d77fe5454=true; Ratings=null; _pxvid=f6abe72c-49a6-11ec-b844-57447254714c; ddbSiteBanner:fd981371-c501-4638-a64b-0d9a5c8da68b=true; Geo={%22region%22:%22ON%22%2C%22country%22:%22CA%22%2C%22continent%22:%22NA%22}; RequestVerificationToken=983cd9ce-c03c-4b8b-b6c1-a0561b39cb47; sublevel=ANON; Preferences=undefined; User.ID=109926924; User.Username=squeaky2x3; UserInfo={"UserId":109926924,"UserJoinDate":"2021-03-25","UserSessionId":"77896779-c672-42d1-903b-f6caba6ec599"}; pxcts=5eb1b230-525e-11ec-aa22-793bf1473016; WarningNotification.Lock=1' \
-      -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; rv:91.0) Gecko/20100101 Firefox/91.0' \
-      -H 'Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8' \
-      -H 'Accept-Language: en-US,en;q=0.5' \
-      -H 'Referer: https://www.dndbeyond.com/' \
-      -H 'DNT: 1' \
-      -H 'Connection: keep-alive' \
-      -H 'Upgrade-Insecure-Requests: 1' \
-      -H 'Sec-Fetch-Dest: document' \
-      -H 'Sec-Fetch-Mode: navigate' \
-      -H 'Sec-Fetch-Site: same-origin' \
-      -H 'Sec-Fetch-User: ?1' \
-      -H 'Pragma: no-cache' \
-      -H 'Cache-Control: no-cache' \
-      -H 'TE: trailers';
 }

 url_to_title_candidate() {
@@ -238,35 +237,79 @@ url_to_title_candidate() {
 html_to_markdown() {
   local f="$(mktemp)"
   log url=$1
-   log 1: url=$1
   cat > "$f"
-   log 2: url=$1
-   #grep '<title>' "$f" | sed 's/.*>\([^<]*\)<.title>.*/# \1/g'
+         #| perl -pe 's|<div[^>]*stat-block-ability-scores-data[^>]*>(.*?)</div>|<span>\1</span>|g' \
+         #| perl -pe 's|<div[^>]*stat-block-ability-scores-heading[^>]*>(.*?)</div>|<strong>\1</strong>|g' \
+         #| perl -pe 's|<div[^>]*stat-block-ability-scores-stat[^>]*>(.*?)</div>|<p>\1</p>|g' \
+      #| readability "file://$(realpath $f)" \
+
   local clean="$(
      cat "$f" \
-      | readability "file://$(realpath $f)" \
-      | pandoc -f html -t markdown --ascii --toc --wrap=none --strip-comments -o - \
+      | awk '/CONTENT/,/FOOTER/' \
+      | tr '\n' '\r' \
+         | perl -pe 's|<blockquote[^>]*>(.*?)</blockquote>|\1|g' \
+         | perl -pe 's|\r<p[^\r]*<strong>([^<]*)</strong></p>\r|\r<h5>\1</h5>\r|g' \
+         | perl -pe 's|<t(.)><p>(.*?)</p></t.>|<t\1>\2</t\1>|g' \
+         | perl -pe 's|<thead>\r*<tr>(?:(?!</tr>).)+</tr>\r*<tr>|<thead><tr>|g' \
+      | tr '\r' '\n' \
+      | sed 's/<br>/ /g' \
+      | sed 's/colspan="[^"]*"//g' \
+      | tr '\n' '\r' \
+         | perl -pe 's|<t(.)><p>(.*?)</p></t.>|<t\1>\2</t\1>|g' \
+         | perl -pe 's|<thead>\r*<tr>(?:(?!</tr>).)+</tr>\r*<tr>|<thead><tr>|g' \
+      | tr '\r' '\n' \
+      | pandoc \
+         -f html \
+         -t markdown+pipe_tables-simple_tables-multiline_tables  \
+         --atx-headers \
+         --ascii \
+         --toc \
+         --wrap=none \
+         --strip-comments \
+         -o - \
+      | grep -v '^\+[-+]*\+$' \
      | sed 's/{[^}]*data[^}]*}//g' \
-      | sed 's/\[](#[^)]*)//'
+      | sed 's/\[](#[^)]*)//' \
+      | cat \
+      | sed 's/{[^}]*}//' \
+      | sed 's/ *$//' \
+      | tr '\n' '\r' \
+         | sed 's/<div>\(.*\)<\/div>/\1/g' \
+      | tr '\r' '\n' \
   )"
-   log 3: url=$1
+   (
+      grep -o '<title.*' "$f" \
+         | sed 's/^[^>]*>//' \
+         | sed 's/-.*//' \
+         | sed 's/^/# /' \
+         | grep . || echo "# ${1##*/}"
+      echo
+      echo "$clean"
+   ) \
+      | tr '\n' '\r' \
+         | sed 's/\r\r:::\r\r/\r/g' \
+         | sed 's/\r\r:::\r*$//' \
+         | sed 's/::://g' \
+         | sed 's/&.dquo;/"/g' \
+         | sed "s/&.squo;/'/g" \
+         | sed 's/&mdash;/--/g' \
+         | sed 's/&ndash;/-/g' \
+         | sed 's/&times;/*/g' \
+         | sed 's/&shy;//g' \
+         | sed 's/{\.[^}]*}//g' \
+      | tr '\r' '\n' \
+      | sed 's/===+/===|/g' \
+      | sed 's/+===/|===/g' \
+      | sed 's/---+/---|/g' \
+      | sed 's/+---/|---/g' \
+      | sed '/^|[=|]*|$/s/=/-/g' \
+
+      #| grep -v '^|-[-|]*-|$' \
+         #| sed '/ |\r|-[-|]*-|\r| /s/-/=/g' \
+         #| sed 's/\r\r| \([^\r]*\) |\r|[^=]/\r\r| \1 |\r|=============================|\r|/g' \
+         #| sed 's/:::\(.*\):::/```\1```/g' \
+
   rm "$f"
-   log 4: url=$1
-   echo "$clean" | (
-      lastline=""
-      while read -r line; do
-         if [ "$line" != "${line#----}" ]; then
-            echo "# $lastline"
-            echo ""
-            break
-         fi
-         lastline="$line"
-      done
-      cat &> /dev/null
-   )
-   log 5: url=$1
-   echo "$clean"
-   log 6: url=$1
   return $?
 }

--- a/to_meili.sh
+++ b/to_meili.sh
@@ -31,24 +31,41 @@ clean_id() {
   echo "$*" | sed 's/[^a-zA-Z0-9_-]/_/g'
 }

+spaceless() {
+   local input="$*"
+   if [ "$#" -lt 1 ]; then
+      input="$(cat)"
+   fi
+   echo "$input" | sed 's/^ *//' | sed 's/ *$//'
+}
+
 scraped_to_meili() {
   find "$SCRAPED" -type f \
   | sort \
   | while read -r md_path; do
-      file_id="$(clean_id "$md_path")"
-      h1="$(head -n 1 "$md_path" | sed 's/^# //')"
-      grep '^## ' "$md_path" | sed 's/^## //' | while read -r h2; do
+      h1="$(grep '^#' "$md_path" | head -n 1 | sed 's/^##*//' | spaceless)"
+      grep '^##*' "$md_path" | while read -r h2_line; do
+         local h2="$(echo "$h2_line" | sed 's/^##*//' | spaceless)"
         local content="$(
            echo "# $h1"
-            echo
+            echo ""
            echo "## $h2"
-            echo
-            cat "$md_path" \
-            | sed -e '0,/^\#\# '"$h2"'/d' \
-            | sed -e '/^\#\# .*/,$d' \
-            | head -n 25
+            echo ""
+            found_h2=false
+            found_stopper=false
+            cat "$md_path" | while read -r line; do
+               if ! $found_h2 && [ "$line" == "$h2_line" ]; then
+                  found_h2=true
+               elif $found_h2 && [ "$line" != "${line#"#"}" ]; then
+                  found_stopper=true
+               fi
+               if $found_stopper; then continue; fi
+               if $found_h2; then echo "$line"; fi
+            done
         )"
-         if [ $(echo "$content" | wc -l | awk '{print $NF}') -lt 5 ]; then
+         content="$(echo "$content" | head -n 25)"
+         if [ $(echo "$content" | wc -w | awk '{print $NF}') -lt 25 ]; then
+            log "skipping content because less than 25 words found: $content"
            continue
         fi
         local id="$(clean_id "${file_id}_${h1}_${h2}")"
@@ -56,6 +73,7 @@ scraped_to_meili() {
         log h1=$h1
         log h2=$h2
         log content="${#content}"
+         log "submitting $id/$h1/$h2"
         curl -sS \
            "$MEILI"/indexes/scraped/documents \
            -X POST \
@@ -69,3 +87,11 @@ scraped_to_meili() {
      done
   done
 }
+
+if [ "$#" == 0 ] ; then
+   if (read -p "would you like to ingest from $SCRAPED? " yn; echo "$yn" | grep -q -i y); then
+      scraped_to_meili
+      echo $?
+   fi
+
+fi
Author	SHA1	Message	Date
bel	882de95140	pbh actually looks good including tables	2022-07-04 11:45:55 -06:00
bel	d6df61c282	tables without thead dont render	2022-07-04 11:27:44 -06:00
bel	cfa69a88fc	cleanup table displays	2022-07-04 10:25:18 -06:00
bel	b445441b84	fix	2022-07-03 23:54:43 -06:00
bel	0977f17c5e	scrape a little bit cleaner about pre-purging lines, tables as preformatted, rm excess :::	2022-07-03 23:31:59 -06:00
bel	f878486d80	update to meili for new scrape format	2022-07-03 21:18:20 -06:00
bel	260815e53c	rm logs	2022-06-27 19:25:17 -06:00
bel	69a2003398	fix illegal chars in title	2022-06-27 19:24:55 -06:00
bel	7c06268f8e	readme	2022-06-27 19:06:43 -06:00