summaryrefslogtreecommitdiff
path: root/bin/crashreportScraper.py
blob: 54477f6c4c284d7984fdc61b7071394ece80b811 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
#!/usr/bin/env python3

# This file is part of the LibreOffice project.
#
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.

# Use this script to retrieve information from https://crashreport.libreoffice.org
# about a specific version of LibreOffice
# Usage sample: ./crashreportScraper.py --version 7.2.0.4 --repository /path/to/libreoffice/repository/

import argparse
import requests
from bs4 import BeautifulSoup
import sys
import os
import math
from datetime import datetime
import urllib.parse

def convert_str_to_date(value):
    value = value.replace('.', '')
    value = value.replace('March', 'Mar')
    value = value.replace('April', 'Apr')
    value = value.replace('June', 'Jun')
    value = value.replace('July', 'Jul')
    value = value.replace('Sept', 'Sep')
    # reset the time leaving the date
    value = ", ".join(value.split(", ")[:-1])
    dtDate = datetime.strptime(value, '%b %d, %Y')

    return dtDate.strftime('%y/%m/%d')

def parse_version_url(url):
    crashReports = {}

    try:
        html_text = requests.get(url, timeout=200).text
        soup = BeautifulSoup(html_text, 'html.parser')
    except requests.exceptions.Timeout:
        print("Timeout requesting " + url)
        sys.exit(1)

    table = soup.find("table", {"id": "data-table"}).tbody
    for tr in table.find_all("tr"):
        td_list = tr.find_all("td")
        crashName = td_list[0].a.text.strip()
        crashNumber = int(td_list[1].text.strip())
        firstCrashDate = convert_str_to_date(td_list[5].text.strip())
        lastCrashDate = convert_str_to_date(td_list[6].text.strip())
        crashReports[crashName] = [crashNumber, firstCrashDate, lastCrashDate]

    return crashReports

def parse_reports_and_get_most_recent_report_from_last_page(url):
    try:
        html_text = requests.get(url, timeout=200).text
        soup = BeautifulSoup(html_text, 'html.parser')
    except requests.exceptions.Timeout:
        print("Timeout")
        raise

    count = 0
    try:
        os_tab = soup.find("table", {"id": "os_tab"}).tbody
    except AttributeError:
        print("os_tab not found")
        raise

    tr_list = os_tab.find_all("tr")
    for tr in tr_list:
        td_list = tr.find_all("td")
        count += int(td_list[1].text.strip())

    # There are 50 reports on each page.
    # Go to the last page based on the total count to get a recent report
    last_page = math.ceil( count / 50 )

    if last_page > 1:
        url = url + "?page=" + str(last_page)
        try:
            html_text = requests.get(url, timeout=200).text
            soup = BeautifulSoup(html_text, 'html.parser')
        except requests.exceptions.Timeout:
            print("Timeout")
            raise

    reports = soup.find("div", {"id": "reports"}).tbody
    ID, currentID = "", ""
    version, currentVersion = "", ""
    OS, currentOS = "", ""

    tr_list = reports.find_all("tr")
    for tr in tr_list:
        td_list = tr.find_all("td")

        currentID = td_list[0].a.text.strip()
        currentVersion = td_list[2].text.strip().split(': ')[1]
        currentOS = td_list[3].text.strip()

        # get most recent version
        # symbols on linux are not very informative generally
        if currentOS == "windows" and currentVersion > version:
            version = currentVersion
            ID = currentID
            OS = currentOS

    if not version:
        version = currentVersion

    if not ID:
        ID = currentID

    if not OS:
        OS = currentOS

    return count, ID, version, OS

def parse_details_and_get_info(url, gitRepo):
    try:
        html_text = requests.get(url, timeout=200).text
        soup = BeautifulSoup(html_text, 'html.parser')
    except requests.exceptions.Timeout:
        print("Timeout")
        raise

    details = soup.find("div", {"id": "details"}).tbody
    tr_list = details.find_all("tr")
    reason = tr_list[8].td.text.strip()

    stack = ""
    codeLine = ""

    count = 0
    frames = soup.find("div", {"id": "frames"}).tbody
    for tr in frames.find_all("tr"):
        td_list = tr.find_all("td")
        source = td_list[3].text.strip()
        if source and count <= 10:
            source = source.replace("\\", "/").replace("C:/cygwin64/home/buildslave/source/libo-core/", "")
            stack += source + "\n"
            count += 1

            codeFile = source.split(":")[0]
            codeNumber = source.split(":")[1]
            try:
                with open(os.path.join(gitRepo, codeFile)) as f:
                    lines = f.readlines()
                    for index, line in enumerate(lines):
                        if index + 1 == int(codeNumber):
                            codeLine += line.strip().replace("\"", "'") + "\n"
            except FileNotFoundError:
                codeLine += "\n"
                continue

    if stack:
        #multiline
        stack = "\"" + stack + "\""

    if codeLine:
        #multiline
        codeLine = "\"" + codeLine + "\""

    metadata = soup.find("div", {"id": "metadata"}).tbody
    tr_list = metadata.find_all("tr")
    unoCommands = ""
    for tr in tr_list:
        if tr.th.text.strip() == "Last-4-Uno-Commands":
            unoCommands = tr.td.text.strip()

    return reason, stack, codeLine, unoCommands


if __name__ == '__main__':

    parser = argparse.ArgumentParser()

    parser.add_argument('--version', action='store', dest="version", required=True)
    parser.add_argument('--repository', action="store", dest="repository", required=True)

    args = parser.parse_args()

    crashes = parse_version_url(
            "https://crashreport.libreoffice.org/stats/version/" + args.version + "?limit=1000&days=30")

    print(str(len(crashes)) + " crash reports in version " + args.version)

    crashesInFile = []
    fileName = "crashes_" + args.version.replace(".", "_") + ".csv"
    print("Using " + fileName)

    bInsertHeader = False
    if os.path.exists(fileName):
        with open(fileName, "r") as f:
            lines = f.readlines()
            for line in lines:
                crashesInFile.append(line.split("\t")[0])
    else:
        bInsertHeader = True

    with open(fileName, "a") as f:
        if bInsertHeader:
            line = '\t'.join(["Name", "Count", "First report", "Last Report",
                "ID", "Version", "Reason", "OS", "Stack", "Code Lines", "Last 4 UNO Commands", '\n'])
            f.write(line)
            f.flush()

        for k, lDate in crashes.items():
            if k not in crashesInFile:
                print("Parsing " + k)
                try:
                    crashCount, crashID, crashVersion, crashOS = parse_reports_and_get_most_recent_report_from_last_page(
                            "https://crashreport.libreoffice.org/stats/signature/" + urllib.parse.quote(k))
                    crashReason, crashStack, codeLine, unoCommands = parse_details_and_get_info(
                            "https://crashreport.libreoffice.org/stats/crash_details/" + crashID, args.repository)
                    line = '\t'.join([k, str(crashCount), lDate[1], lDate[2],
                            crashID, crashVersion, crashReason, crashOS, crashStack, codeLine, unoCommands, '\n'])
                    f.write(line)
                    f.flush()
                except (requests.exceptions.Timeout, AttributeError):
                    continue