[PATCH 2/2] scripts/contrib: add oe-build-perf-report-email

[Markus Lehtonen <markus.lehtonen@linux.intel.com>]

Script for sending build perf test reports as an email. Mangles an html
report, generated by oe-build-perf-report, into a format suitable for
html emails. Supports multipart emails where a plaintext alternative can
be included in the same email.

Dependencies required to be installed on the host:
- phantomjs
- optipng

[YOCTO #10931]

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
---
 scripts/contrib/oe-build-perf-report-email.py | 266 ++++++++++++++++++++++++++
 scripts/lib/build_perf/scrape-html-report.js  |  56 ++++++
 2 files changed, 322 insertions(+)
 create mode 100755 scripts/contrib/oe-build-perf-report-email.py
 create mode 100644 scripts/lib/build_perf/scrape-html-report.js

diff --git a/scripts/contrib/oe-build-perf-report-email.py b/scripts/contrib/oe-build-perf-report-email.py
new file mode 100755
index 0000000..7f4274e
--- /dev/null
+++ b/scripts/contrib/oe-build-perf-report-email.py
@@ -0,0 +1,266 @@
+#!/usr/bin/python3
+#
+# Send build performance test report emails
+#
+# Copyright (c) 2017, Intel Corporation.
+#
+# This program is free software; you can redistribute it and/or modify it
+# under the terms and conditions of the GNU General Public License,
+# version 2, as published by the Free Software Foundation.
+#
+# This program is distributed in the hope it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+# more details.
+#
+import argparse
+import base64
+import logging
+import os
+import pwd
+import re
+import shutil
+import smtplib
+import subprocess
+import sys
+import tempfile
+from email.mime.multipart import MIMEMultipart
+from email.mime.text import MIMEText
+
+
+# Setup logging
+logging.basicConfig(level=logging.INFO, format="%(levelname)s: %(message)s")
+log = logging.getLogger('oe-build-perf-report')
+
+
+# Find js scaper script
+SCRAPE_JS = os.path.join(os.path.dirname(__file__), '..', 'lib', 'build_perf',
+                         'scrape-html-report.js')
+if not os.path.isfile(SCRAPE_JS):
+    log.error("Unableto find oe-build-perf-report-scrape.js")
+    sys.exit(1)
+
+
+class ReportError(Exception):
+    """Local errors"""
+    pass
+
+
+def check_utils():
+    """Check that all needed utils are installed in the system"""
+    missing = []
+    for cmd in ('phantomjs', 'optipng'):
+        if not shutil.which(cmd):
+            missing.append(cmd)
+    if missing:
+        log.error("The following tools are missing: %s", ' '.join(missing))
+        sys.exit(1)
+
+
+def parse_args(argv):
+    """Parse command line arguments"""
+    description = """Email build perf test report"""
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
+        description=description)
+
+    parser.add_argument('--debug', '-d', action='store_true',
+                        help="Verbose logging")
+    parser.add_argument('--quiet', '-q', action='store_true',
+                        help="Only print errors")
+    parser.add_argument('--to', action='append',
+                        help="Recipients of the email")
+    parser.add_argument('--subject', default="Yocto build perf test report",
+                        help="Email subject")
+    parser.add_argument('--outdir', '-o',
+                        help="Store files in OUTDIR. Can be used to preserve "
+                             "the email parts")
+    parser.add_argument('--text',
+                        help="Plain text message")
+    parser.add_argument('--html',
+                        help="HTML peport generated by oe-build-perf-report")
+    parser.add_argument('--phantomjs-args', action='append',
+                        help="Extra command line arguments passed to PhantomJS")
+
+    args = parser.parse_args(argv)
+
+    if not args.html and not args.text:
+        parser.error("Please specify --html and/or --text")
+
+    return args
+
+
+def decode_png(infile, outfile):
+    """Parse/decode/optimize png data from a html element"""
+    with open(infile) as f:
+        raw_data = f.read()
+
+    # Grab raw base64 data
+    b64_data = re.sub('^.*href="data:image/png;base64,', '', raw_data, 1)
+    b64_data = re.sub('">.+$', '', b64_data, 1)
+
+    # Replace file with proper decoded png
+    with open(outfile, 'wb') as f:
+        f.write(base64.b64decode(b64_data))
+
+    subprocess.check_output(['optipng', outfile], stderr=subprocess.STDOUT)
+
+
+def encode_png(pngfile):
+    """Encode png into a <img> html element"""
+    with open(pngfile, 'rb') as f:
+        data = f.read()
+
+    b64_data = base64.b64encode(data)
+    return '<img src="data:image/png;base64,' + b64_data.decode('utf-8') + '">\n'
+
+
+def mangle_html_report(infile, outfile, pngs):
+    """Mangle html file into a email compatible format"""
+    paste = True
+    png_dir = os.path.dirname(outfile)
+    with open(infile) as f_in:
+        with open(outfile, 'w') as f_out:
+            for line in f_in.readlines():
+                stripped = line.strip()
+                # Strip out scripts
+                if stripped == '<!--START-OF-SCRIPTS-->':
+                    paste = False
+                elif stripped == '<!--END-OF-SCRIPTS-->':
+                    paste = True
+                elif paste:
+                    if re.match('^.+href="data:image/png;base64', stripped):
+                        # Strip out encoded pngs (as they're huge in size)
+                        continue
+                    elif 'www.gstatic.com' in stripped:
+                        # HACK: drop references to external static pages
+                        continue
+
+                    # Replace charts with <img> elements
+                    match = re.match('<div id="(?P<id>\w+)"', stripped)
+                    if match and match.group('id') in pngs:
+                        #f_out.write('<img src="{}">\n'.format(match.group('id') + '.png'))
+                        png_file = os.path.join(png_dir, match.group('id') + '.png')
+                        f_out.write(encode_png(png_file))
+                    else:
+                        f_out.write(line)
+
+
+def scrape_html_report(report, outdir, phantomjs_extra_args=None):
+    """Scrape html report into a format sendable by email"""
+    tmpdir = tempfile.mkdtemp(dir='.')
+    log.debug("Using tmpdir %s for phantomjs output", tmpdir)
+
+    if not os.path.isdir(outdir):
+        os.mkdir(outdir)
+    if os.path.splitext(report)[1] not in ('.html', '.htm'):
+        raise ReportError("Invalid file extension for report, needs to be "
+                          "'.html' or '.htm'")
+
+    try:
+        log.info("Scraping HTML report with PhangomJS")
+        extra_args = phantomjs_extra_args if phantomjs_extra_args else []
+        subprocess.check_output(['phantomjs', '--debug=true'] + extra_args +
+                                [SCRAPE_JS, report, tmpdir],
+                                stderr=subprocess.STDOUT)
+
+        pngs = []
+        attachments = []
+        for fname in os.listdir(tmpdir):
+            base, ext = os.path.splitext(fname)
+            if ext == '.png':
+                log.debug("Decoding %s", fname)
+                decode_png(os.path.join(tmpdir, fname),
+                           os.path.join(outdir, fname))
+                pngs.append(base)
+                attachments.append(fname)
+            elif ext in ('.html', '.htm'):
+                report_file = fname
+            else:
+                log.warning("Unknown file extension: '%s'", ext)
+                #shutil.move(os.path.join(tmpdir, fname), outdir)
+
+        log.debug("Mangling html report file %s", report_file)
+        mangle_html_report(os.path.join(tmpdir, report_file),
+                           os.path.join(outdir, report_file), pngs)
+        return report_file, attachments
+    finally:
+        shutil.rmtree(tmpdir)
+
+def send_email(text_fn, html_fn, subject, recipients):
+    """Send email"""
+    # Generate email message
+    text_msg = html_msg = None
+    if text_fn:
+        with open(text_fn) as f:
+            text_msg = MIMEText("Yocto build performance test report.\n" +
+                                f.read(), 'plain')
+    if html_fn:
+        with open(html_fn) as f:
+            html_msg = MIMEText(f.read(), 'html')
+
+    if text_msg and html_msg:
+        msg = MIMEMultipart('alternative')
+        msg.attach(text_msg)
+        msg.attach(html_msg)
+    elif text_msg:
+        msg = text_msg
+    elif html_msg:
+        msg = html_msg
+    else:
+        raise ReportError("Neither plain text nor html body specified")
+
+    full_name = pwd.getpwuid(os.getuid()).pw_gecos.split(',')[0]
+    email = os.environ.get('EMAIL', os.getlogin())
+    msg['From'] = "{} <{}>".format(full_name, email)
+    msg['To'] = ', '.join(recipients)
+    msg['Subject'] = subject
+
+    # Send email
+    with smtplib.SMTP('localhost') as smtp:
+        smtp.send_message(msg)
+
+
+def main(argv=None):
+    """Script entry point"""
+    args = parse_args(argv)
+    if args.quiet:
+        log.setLevel(logging.ERROR)
+    if args.debug:
+        log.setLevel(logging.DEBUG)
+
+    check_utils()
+
+    if args.outdir:
+        outdir = args.outdir
+        if not os.path.exists(outdir):
+            os.mkdir(outdir)
+    else:
+        outdir = tempfile.mkdtemp(dir='.')
+
+    try:
+        log.debug("Storing email parts in %s", outdir)
+        html_report = None
+        if args.html:
+            scrape_html_report(args.html, outdir, args.phantomjs_args)
+            html_report = os.path.join(outdir, args.html)
+
+        if args.to:
+            log.info("Sending email to %s", ', '.join(args.to))
+            send_email(args.text, html_report, args.subject, args.to)
+    except subprocess.CalledProcessError as err:
+        log.error("%s, with output:\n%s", str(err), err.output.decode())
+        return 1
+    except ReportError as err:
+        log.error(err)
+        return 1
+    finally:
+        if not args.outdir:
+            log.debug("Wiping %s", outdir)
+            shutil.rmtree(outdir)
+
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/scripts/lib/build_perf/scrape-html-report.js b/scripts/lib/build_perf/scrape-html-report.js
new file mode 100644
index 0000000..05a1f57
--- /dev/null
+++ b/scripts/lib/build_perf/scrape-html-report.js
@@ -0,0 +1,56 @@
+var fs = require('fs');
+var system = require('system');
+var page = require('webpage').create();
+
+// Examine console log for message from chart drawing
+page.onConsoleMessage = function(msg) {
+    console.log(msg);
+    if (msg === "ALL CHARTS READY") {
+        window.charts_ready = true;
+    }
+    else if (msg.slice(0, 11) === "CHART READY") {
+        var chart_id = msg.split(" ")[2];
+        console.log('grabbing ' + chart_id);
+        var png_data = page.evaluate(function (chart_id) {
+            var chart_div = document.getElementById(chart_id + '_png');
+            return chart_div.outerHTML;
+        }, chart_id);
+        fs.write(args[2] + '/' + chart_id + '.png', png_data, 'w');
+    }
+};
+
+// Check command line arguments
+var args = system.args;
+if (args.length != 3) {
+    console.log("USAGE: " + args[0] + " REPORT_HTML OUT_DIR\n");
+    phantom.exit(1);
+}
+
+// Open the web page
+page.open(args[1], function(status) {
+    if (status == 'fail') {
+        console.log("Failed to open file '" + args[1] + "'");
+        phantom.exit(1);
+    }
+});
+
+// Check status every 100 ms
+interval = window.setInterval(function () {
+    //console.log('waiting');
+    if (window.charts_ready) {
+        clearTimeout(timer);
+        clearInterval(interval);
+
+        var fname = args[1].replace(/\/+$/, "").split("/").pop()
+        console.log("saving " + fname);
+        fs.write(args[2] + '/' + fname, page.content, 'w');
+        phantom.exit(0);
+    }
+}, 100);
+
+// Time-out after 10 seconds
+timer = window.setTimeout(function () {
+    clearInterval(interval);
+    console.log("ERROR: timeout");
+    phantom.exit(1);
+}, 10000);
-- 
2.10.2