Re: [RFC v2 2/3] recheck: Add a recheck parser for patchwork comments

[Michael Santana <msantana@redhat.com>]

On Tue, Nov 7, 2023 at 3:32 PM Aaron Conole <aconole@redhat.com> wrote:
>
> Add a recheck parsing tool that will allow for labs to build a
> recheck workflow based on specific recheck labels and projects,
> with an associated state machine and querying interface.
>
> The output of the recheck parsing tool is json and can be fed to
> jq or other json parsing utilities for better field support.
>
> Signed-off-by: Aaron Conole <aconole@redhat.com>
Acked-by: Michael Santana <msantana@redhat.com>
> ---
>  pw_mon           |  58 ++++++++++++++++++++++++++-
>  recheck_tool     | 100 +++++++++++++++++++++++++++++++++++++++++++++++
>  series_db_lib.sh |  64 +++++++++++++++++++++++++++++-
>  3 files changed, 220 insertions(+), 2 deletions(-)
>  create mode 100755 recheck_tool
>
> diff --git a/pw_mon b/pw_mon
> index da4b9a9..29a9cb4 100755
> --- a/pw_mon
> +++ b/pw_mon
> @@ -1,6 +1,6 @@
>  #!/bin/sh
>  # SPDX-Identifier: gpl-2.0-or-later
> -# Copyright (C) 2018, Red Hat, Inc.
> +# Copyright (C) 2018-2023 Red Hat, Inc.
>  #
>  # Monitors a project on a patchwork instance for new series submissions
>  # Records the submissions in the series database (and emits them on the
> @@ -44,6 +44,7 @@ if [ "$1" != "" ]; then
>      fi
>  fi
>
> +recheck_filter=""
>
>  while [ "$1" != "" ]; do
>      if echo "$1" | grep -q -s -E ^--pw-project= ; then
> @@ -67,6 +68,10 @@ while [ "$1" != "" ]; do
>          echo "    --add-filter-recheck=filter  Adds a filter to flag that a recheck needs to be done"
>          echo ""
>          exit 0
> +    elif echo "$1" | grep -q -s -E ^--add-filter-recheck=; then
> +        filter_str=$(echo "$1" | sed s/--add-filter-recheck=//)
> +        recheck_filter="$filter_str $recheck_filter"
> +        shift
>      else
>          echo "Unknown option: '$1'"
>          echo "Rerun with --help for details"
> @@ -191,7 +196,58 @@ function check_superseded_series() {
>      done
>  }
>
> +function run_recheck() {
> +    local recheck_list=$(echo "$7" | sed -e 's/^Recheck-request: // ' -e 's/,/ /g')
> +
> +    for filter in $recheck_filter; do
> +        for check in $recheck_list; do
> +            if [ "$filter" == "$check" ]; then
> +                insert_recheck_request_if_needed "$1" "$3" "$8" "$check" "$2" "$9"
> +            fi
> +        done
> +    done
> +}
> +
> +function check_patch_for_retest_request() {
> +    local patch_url="$1"
> +
> +    local patch_json=$(curl -s "$userpw" "$patch_url")
> +    local patch_comments_url=$(echo "$patch_json" | jq -rc '.comments')
> +    local patch_id=$(echo "$patch_json" | jq -rc '.id')
> +    if [ "Xnull" != "X$patch_comments_url" ]; then
> +        local comments_json=$(curl -s "$userpw" "$patch_comments_url")
> +
> +        local seq_end=$(echo "$comments_json" | jq -rc 'length')
> +        if [ "$seq_end" -a $seq_end -gt 0 ]; then
> +            seq_end=$((seq_end-1))
> +            for comment_id in $(seq 0 $seq_end); do
> +                local recheck_requested=$(echo "$comments_json" | jq -rc ".[$comment_id].content" | grep "^Recheck-request: ")
> +                if [ "X$recheck_requested" != "X" ]; then
> +                    local msgid=$(echo "$comments_json" | jq -rc ".[$comment_id].msgid")
> +                    run_recheck "$pw_instance" "$series_id" "$project" "$url" "$repo" "$branchname" "$recheck_requested" "$msgid" "$patch_id"
> +                fi
> +            done
> +        fi
> +    fi
> +}
> +
> +function check_series_needs_retest() {
> +    local pw_instance="$1"
> +
> +    series_get_active_branches "$pw_instance" | while IFS=\| read -r series_id project url repo branchname; do
> +        local patch_comments_url=$(curl -s "$userpw" "$url" | jq -rc '.patches[] | .url')
> +
> +        for patch in $patch_comments_url; do
> +            check_patch_for_retest_request $patch
> +        done
> +    done
> +}
> +
>  check_undownloaded_series "$pw_instance" "$pw_project"
>  check_completed_series "$pw_instance" "$pw_project"
>  check_new_series "$pw_instance" "$pw_project"
>  check_superseded_series "$pw_instance"
> +
> +# check for retest requests after a series is still passing all the
> +# checks above
> +check_series_needs_retest "$pw_instance"
> diff --git a/recheck_tool b/recheck_tool
> new file mode 100755
> index 0000000..98031a9
> --- /dev/null
> +++ b/recheck_tool
> @@ -0,0 +1,100 @@
> +#!/bin/sh
> +# SPDX-Identifier: gpl-2.0-or-later
> +# Copyright (C) 2023 Red Hat, Inc.
> +#
> +# Licensed under the terms of the GNU General Public License as published
> +# by the Free Software Foundation; either version 2 of the License, or
> +# (at your option) any later version.  You may obtain a copy of the
> +# license at
> +#
> +#    https://www.gnu.org/licenses/old-licenses/gpl-2.0.html
> +#
> +# Unless required by applicable law or agreed to in writing, software
> +# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
> +# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
> +# License for the specific language governing permissions and limitations
> +# under the License.
> +
> +mode="select"
> +
> +while [ "$1" != "" ]; do
> +    if echo "$1" | grep -q -s -E ^--help ; then
> +        echo "recheck / retest state machine script"
> +        echo ""
> +        echo "$0:"
> +        echo " --pw-project=<proj>:    Patchwork project."
> +        echo " --pw-instance=<inst>:   Patchwork instance."
> +        echo " --filter=<str>:         Job / request for recheck."
> +        echo " --state=<0..>:          Resync state ID."
> +        echo " --msgid=<msgid>:                Message ID to select."
> +        echo " --update:               Set tool in update mode"
> +        echo " --new-state=<0..>:      New state ID to set"
> +        echo " --series-id=<..>:       Series ID"
> +        echo ""
> +        echo "Will spit out a parsable json for each db line when selecting"
LGTM. But to Dumitru's point, maybe we should add documentation this
this in a follow up patch
> +        exit 0
> +    elif echo "$1" | grep -q -s -E ^--pw-project= ; then
> +        pw_project=$(echo "$1" | sed s/^--pw-project=//)
> +    elif echo "$1" | grep -q -s -E ^--pw-instance= ; then
> +        pw_instance=$(echo "$1" | sed s/^--pw-instance=//)
> +    elif echo "$1" | grep -q -s -E ^--filter= ; then
> +        filter=$(echo "$1" | sed s/^--filter=//)
> +    elif echo "$1" | grep -q -s -E ^--state= ; then
> +        recheck_state=$(echo "$1" | sed s/^--state=//)
> +    elif echo "$1" | grep -q -s -E ^--msgid= ; then
> +        message_id=$(echo "$1" | sed s/^--msgid=//)
> +    elif echo "$1" | grep -q -s -E ^--update ; then
> +        mode="update"
> +    elif echo "$1" | grep -q -s -E ^--new-state= ; then
> +        new_recheck_state=$(echo "$1" | sed s/^--new-state=//)
> +    elif echo "$1" | grep -q -s -E ^--series-id= ; then
> +        series_id=$(echo "$1" | sed s/^--series-id=//)
> +    else
> +        echo "Unknown option: '$1'"
> +        echo "Rerun with --help for details"
> +        exit 1
> +    fi
> +    shift
> +done
> +
> +source $(dirname $0)/series_db_lib.sh
> +
> +if [ "$mode" == "select" ]; then
> +    echo -n "{\"rechecks\":["
> +    printed=""
> +    for request in $(get_recheck_requests_by_project "$pw_instance" \
> +                                                     "$pw_project" \
> +                                                     "$recheck_state" \
> +                                                     "$filter"); do
> +        message_id=$(echo $request | cut -d\| -f1)
> +        series_id=$(echo $request | cut -d\| -f2)
> +        patch_id=$(echo $request | cut -d\| -f3)
> +
> +        sha=$(get_sha_for_series_id_and_patch "$series_id" "$patch_id" \
> +                                              "$pw_instance")
> +        echo -n "$printed{\"pw_instance\":\"$pw_instance\",\"series_id\":$series_id,\"patch_id\":$patch_id,\"sha\":\"$sha\",\"msg_id\":\"$message_id\",\"state\":\"$recheck_state\",\"requested\":\"$filter\"}"
okay, smart use of the -n for the echo command. I see you are also
taking care of the coma which is good!
> +        printed=","
> +    done
> +    echo "]}"
> +elif [ "$mode" == "update" ]; then
> +    if [ "X$new_recheck_state" == "X" -o "X$series_id" == "X" ]; then
> +        echo "Need to set a series-id and a new recheck state when updating."
> +        exit 1
> +    fi
> +
> +    request=$(get_recheck_request "$pw_instance" "$pw_project" "$message_id" \
> +                                  "$filter" "$series_id" "$recheck_state")
> +    if [ "X$request" == "X" ]; then
> +        echo "{\"result\":\"notfound\"}"
> +        exit 0
> +    fi
> +
> +    set_recheck_request_state "$pw_instance" "$pw_project" "$message_id" \
> +                              "$filter" "$series_id" "$new_recheck_state"
> +
> +    echo "{\"result\":\"executed\",\"recheck\":{\"pw_instance\": \"$pw_instance\", \"series_id\":$series_id, \"msg_id\":\"$message_id\", \"state\":\"$new_recheck_state\", \"requested\": \"$filter\"}}"
okay
> +else
> +    echo "Unknown state: $mode"
> +    exit 1
> +fi
> +
> diff --git a/series_db_lib.sh b/series_db_lib.sh
> index 6c2d98e..ca33c1f 100644
> --- a/series_db_lib.sh
> +++ b/series_db_lib.sh
> @@ -1,6 +1,6 @@
>  #!/bin/sh
>  # SPDX-Identifier: gpl-2.0-or-later
> -# Copyright (C) 2018,2019 Red Hat, Inc.
> +# Copyright (C) 2018-2023 Red Hat, Inc.
>  #
>  # Licensed under the terms of the GNU General Public License as published
>  # by the Free Software Foundation; either version 2 of the License, or
> @@ -114,6 +114,22 @@ EOF
>          run_db_command "INSERT INTO series_schema_version(id) values (7);"
>      fi
>
> +    run_db_command "select * from series_schema_version;" | egrep '^8$' > /dev/null 2>&1
> +    if [ $? -eq 1 ]; then
> +        sqlite3 ${HOME}/.series-db <<EOF
> +CREATE TABLE recheck_requests (
> +recheck_id INTEGER,
> +recheck_message_id STRING,
> +recheck_requested_by STRING,
> +recheck_series STRING,
> +recheck_patch INTEGER,
> +patchwork_instance STRING,
> +patchwork_project STRING,
> +recheck_sync INTEGER
> +);
> +EOF
> +        run_db_command "INSERT INTO series_schema_version(id) values (8);"
> +    fi
>  }
>
>  function series_db_exists() {
> @@ -390,3 +406,49 @@ function get_patch_id_by_series_id_and_sha() {
>
>      echo "select patch_id from git_builds where patchwork_instance=\"$instance\" and series_id=$series_id and sha=\"$sha\";" | series_db_execute
>  }
> +
> +function get_recheck_requests_by_project() {
> +    local recheck_instance="$1"
> +    local recheck_project="$2"
> +    local recheck_state="$3"
> +    local recheck_requested_by="$4"
> +
> +    series_db_exists
> +
> +    echo "select recheck_message_id,recheck_series,recheck_patch from recheck_requests where patchwork_instance=\"$recheck_instance\" and patchwork_project=\"$recheck_project\" and recheck_sync=$recheck_state and recheck_requested_by=\"$recheck_requested_by\";" | series_db_execute
> +}
> +
> +function insert_recheck_request_if_needed() {
> +    local recheck_instance="$1"
> +    local recheck_project="$2"
> +    local recheck_msgid="$3"
> +    local recheck_requested_by="$4"
> +    local recheck_series="$5"
> +    local recheck_patch="$6"
> +
> +    if ! echo "select * from recheck_requests where recheck_message_id=\"$recheck_msgid\";" | series_db_execute | grep $recheck_msgid >/dev/null 2>&1; then
> +        echo "INSERT INTO recheck_requests (recheck_message_id, recheck_requested_by, recheck_series, recheck_patch, patchwork_instance, patchwork_project, recheck_sync) values (\"$recheck_msgid\", \"$recheck_requested_by\", \"$recheck_series\", $recheck_patch, \"$recheck_instance\", \"$recheck_project\", 0);" | series_db_execute
> +    fi
> +}
> +
> +function get_recheck_request() {
> +    local recheck_instance="$1"
> +    local recheck_project="$2"
> +    local recheck_msgid="$3"
> +    local recheck_requested_by="$4"
> +    local recheck_series="$5"
> +    local recheck_state="$6"
> +
> +    echo "select * from recheck_requests where patchwork_instance=\"$recheck_instance\" and patchwork_project=\"$recheck_project\" and recheck_requested_by=\"$recheck_requested_by\" and recheck_series=\"$recheck_series\" and recheck_message_id=\"$recheck_msgid\" and recheck_sync=$recheck_state;" | series_db_execute
> +}
> +
> +function set_recheck_request_state() {
> +    local recheck_instance="$1"
> +    local recheck_project="$2"
> +    local recheck_msgid="$3"
> +    local recheck_requested_by="$4"
> +    local recheck_series="$5"
> +    local recheck_state="$6"
> +
> +    echo "UPDATE recheck_requests set recheck_sync=$recheck_state where patchwork_instance=\"$recheck_instance\" and patchwork_project=\"$recheck_project\" and recheck_requested_by=\"$recheck_requested_by\" and recheck_series=\"$recheck_series\";" | series_db_execute
> +}
> --
> 2.41.0
>