Merge branch upstream/master while introducing compile errors and reverting parts of Android changes, video BPP options are gone

This commit is contained in:
Sergii Pylypenko
2021-11-29 02:17:14 +02:00
667 changed files with 66564 additions and 29169 deletions

View File

@@ -1,17 +0,0 @@
---
name: Bugs
about: Found a bug in OpenTTD?
title: "Bug Report"
---
## Version of OpenTTD
<!-- Fill in below what version of OpenTTD you are using, including your OS. -->
## Expected result
<!-- Describe in a few words what you expected to happen. -->
## Actual result
<!-- Describe in a few words what actually happens. -->
## Steps to reproduce
<!-- As detailed as possible, please tell us how we can reproduce this. Feel free to attach a savegame (zip it first) to make it more clear. -->

41
.github/ISSUE_TEMPLATE/bug.yaml vendored Normal file
View File

@@ -0,0 +1,41 @@
name: Bug Report
description: Found a bug in OpenTTD?
title: "[Bug]: "
body:
- type: markdown
attributes:
value: |
Thanks for taking the time to fill out this bug report!
- type: input
id: version
attributes:
label: Version of OpenTTD
description: Fill in below what version of OpenTTD you are using, including your OS.
placeholder: ex. 1.11.2, Windows 10
validations:
required: true
- type: textarea
id: expected
attributes:
label: Expected result
description: Describe in a few words what you expected to happen.
validations:
required: true
- type: textarea
id: actual
attributes:
label: Actual result
description: Describe in a few words what actually happens.
validations:
required: true
- type: textarea
id: reproduce
attributes:
label: Steps to reproduce
description: As detailed as possible, please tell us how we can reproduce this. Feel free to attach a savegame (zip it first) to make it more clear.
placeholder: |
1. Loaded the attached savegame.
2. Click on the button left of that other icon.
3. The window doesn't open.
validations:
required: true

View File

@@ -1,12 +0,0 @@
---
name: Crash
about: Did OpenTTD crash?
title: "Crash Report"
---
<!-- Please zip the crash.log, crash.dmp and crash.sav and attach it to this crash report. -->
## Version of OpenTTD
<!-- Fill in below what version of OpenTTD you are using, including your OS. -->
## Steps to reproduce
<!-- Please spend a few words if you can reproduce this problem. -->

37
.github/ISSUE_TEMPLATE/crash.yaml vendored Normal file
View File

@@ -0,0 +1,37 @@
name: Crash
description: Did OpenTTD crash?
title: "[Crash]: "
body:
- type: markdown
attributes:
value: |
Thanks for taking the time to fill out this crash report!
- type: input
id: version
attributes:
label: Version of OpenTTD
description: Fill in below what version of OpenTTD you are using, including your OS.
placeholder: ex. 1.11.2, Windows 10
validations:
required: true
- type: textarea
id: reproduce
attributes:
label: Steps to reproduce
description: Please spend a few words if you can reproduce this problem.
placeholder: |
1. Bought a new train.
2. The game crashed.
validations:
required: true
- type: textarea
id: crashlogs
attributes:
label: Upload crash files
description: With the `crash.log`, `crash.dmp`, and `crash.sav` we can analyze the crash in detail; this way you allow us to easier triage and fix the problem.
placeholder: |
1. Zip the `crash.log`, `crash.dmp` and `crash.sav`.
2. Click on this field.
3. Drag and drop the zip file in here.
validations:
required: true

View File

@@ -39,6 +39,7 @@ Describe here
Some things are not automated, and forgotten often. This list is a reminder for the reviewers.
* The bug fix is important enough to be backported? (label: 'backport requested')
* This PR touches english.txt or translations? Check the [guidelines](https://github.com/OpenTTD/OpenTTD/blob/master/docs/eints.md)
* This PR affects the save game format? (label 'savegame upgrade')
* This PR affects the GS/AI API? (label 'needs review: Script API')
* ai_changelog.hpp, gs_changelog.hpp need updating.

View File

@@ -4,9 +4,9 @@ tag=$(git name-rev --name-only --tags --no-undefined HEAD 2>/dev/null | sed 's@\
# If we are a tag, show the part of the changelog till (but excluding) the last stable
if [ -n "$tag" ]; then
grep='^[0-9]\+\.[0-9]\+\.[0-9]\+[^-]'
grep='^[0-9]\+\.[0-9]\+[^-]'
next=$(cat changelog.txt | grep '^[0-9]' | awk 'BEGIN { show="false" } // { if (show=="true") print $0; if ($1=="'$tag'") show="true"} ' | grep "$grep" | head -n1 | sed 's/ .*//')
cat changelog.txt | awk 'BEGIN { show="false" } /^[0-9]+.[0-9]+.[0-9]+/ { if ($1=="'$next'") show="false"; if ($1=="'$tag'") show="true";} // { if (show=="true") print $0 }'
cat changelog.txt | awk 'BEGIN { show="false" } /^[0-9]+.[0-9]+/ { if ($1=="'$next'") show="false"; if ($1=="'$tag'") show="true";} // { if (show=="true") print $0 }'
exit 0
fi

221
.github/unused-strings.py vendored Normal file
View File

@@ -0,0 +1,221 @@
"""
Script to scan the OpenTTD source-tree for STR_ entries that are defined but
no longer used.
This is not completely trivial, as OpenTTD references a lot of strings in
relation to another string. The most obvious example of this is a list. OpenTTD
only references the first entry in the list, and does "+ <var>" to get to the
correct string.
There are other ways OpenTTD does use relative values. This script tries to
account for all of them, to give the best approximation we have for "this
string is unused".
"""
import glob
import os
import re
import subprocess
import sys
from enum import Enum
LENGTH_NAME_LOOKUP = {
"VEHICLE_TYPES": 4,
}
class SkipType(Enum):
NONE = 1
LENGTH = 2
EXTERNAL = 3
ZERO_IS_SPECIAL = 4
EXPECT_NEWLINE = 5
def read_language_file(filename, strings_found, errors):
strings_defined = []
skip = SkipType.NONE
length = 0
common_prefix = ""
last_tiny_string = ""
with open(filename) as fp:
for line in fp.readlines():
if not line.strip():
if skip == SkipType.EXPECT_NEWLINE:
skip = SkipType.NONE
continue
line = line.strip()
if skip == SkipType.EXPECT_NEWLINE:
# The only thing allowed after a list, is this next marker, or a newline.
if line == "###next-name-looks-similar":
# "###next-name-looks-similar"
# Indicates the common prefix of the last list has a very
# similar name to the next entry, but isn't part of the
# list. So do not emit a warning about them looking very
# similar.
if length != 0:
errors.append(f"ERROR: list around {name} is shorted than indicated by ###length")
common_prefix = ""
else:
errors.append(f"ERROR: expected a newline after a list, but didn't find any around {name}. Did you add an entry to the list without increasing the length?")
skip = SkipType.NONE
if line[0] == "#":
if line.startswith("###length "):
# "###length <count>"
# Indicates the next few entries are part of a list. Only
# the first entry is possibly referenced, and the rest are
# indirectly.
if length != 0:
errors.append(f"ERROR: list around {name} is shorted than indicated by ###length")
length = line.split(" ")[1].strip()
if length.isnumeric():
length = int(length)
else:
length = LENGTH_NAME_LOOKUP[length]
skip = SkipType.LENGTH
elif line.startswith("###external "):
# "###external <count>"
# Indicates the next few entries are used outside the
# source and will not be referenced.
if length != 0:
errors.append(f"ERROR: list around {name} is shorted than indicated by ###length")
length = line.split(" ")[1].strip()
length = int(length)
skip = SkipType.EXTERNAL
elif line.startswith("###setting-zero-is-special"):
# "###setting-zero-is-special"
# Indicates the next entry is part of the "zero is special"
# flag of settings. These entries are not referenced
# directly in the code.
if length != 0:
errors.append(f"ERROR: list around {name} is shorted than indicated by ###length")
skip = SkipType.ZERO_IS_SPECIAL
continue
name = line.split(":")[0].strip()
strings_defined.append(name)
# If a string ends on _TINY or _SMALL, it can be the {TINY} variant.
# Check for this by some fuzzy matching.
if name.endswith(("_SMALL", "_TINY")):
last_tiny_string = name
elif last_tiny_string:
matching_name = "_".join(last_tiny_string.split("_")[:-1])
if name == matching_name:
strings_found.add(last_tiny_string)
else:
last_tiny_string = ""
if skip == SkipType.EXTERNAL:
strings_found.add(name)
skip = SkipType.LENGTH
if skip == SkipType.LENGTH:
skip = SkipType.NONE
length -= 1
common_prefix = name
elif skip == SkipType.ZERO_IS_SPECIAL:
strings_found.add(name)
elif length > 0:
strings_found.add(name)
length -= 1
# Find the common prefix of these strings
for i in range(len(common_prefix)):
if common_prefix[0 : i + 1] != name[0 : i + 1]:
common_prefix = common_prefix[0:i]
break
if length == 0:
skip = SkipType.EXPECT_NEWLINE
if len(common_prefix) < 6:
errors.append(f"ERROR: common prefix of block including {name} was reduced to {common_prefix}. This means the names in the list are not consistent.")
elif common_prefix:
if name.startswith(common_prefix):
errors.append(f"ERROR: {name} looks a lot like block above with prefix {common_prefix}. This mostly means that the list length was too short. Use '###next-name-looks-similar' if it is not.")
common_prefix = ""
return strings_defined
def scan_source_files(path, strings_found):
for new_path in glob.glob(f"{path}/*"):
if os.path.isdir(new_path):
scan_source_files(new_path, strings_found)
continue
if not new_path.endswith((".c", ".h", ".cpp", ".hpp", ".ini")):
continue
# Most files we can just open, but some use magic, that requires the
# G++ preprocessor before we can make sense out of it.
if new_path == "src/table/cargo_const.h":
p = subprocess.run(["g++", "-E", new_path], stdout=subprocess.PIPE)
output = p.stdout.decode()
else:
with open(new_path) as fp:
output = fp.read()
# Find all the string references.
matches = re.findall(r"[^A-Z_](STR_[A-Z0-9_]*)", output)
strings_found.update(matches)
def main():
strings_found = set()
errors = []
scan_source_files("src", strings_found)
strings_defined = read_language_file("src/lang/english.txt", strings_found, errors)
# STR_LAST_STRINGID is special, and not really a string.
strings_found.remove("STR_LAST_STRINGID")
# These are mentioned in comments, not really a string.
strings_found.remove("STR_XXX")
strings_found.remove("STR_NEWS")
strings_found.remove("STR_CONTENT_TYPE_")
# This string is added for completion, but never used.
strings_defined.remove("STR_JUST_DATE_SHORT")
strings_defined = sorted(strings_defined)
strings_found = sorted(list(strings_found))
for string in strings_found:
if string not in strings_defined:
errors.append(f"ERROR: {string} found but never defined.")
for string in strings_defined:
if string not in strings_found:
errors.append(f"ERROR: {string} is (possibly) no longer needed.")
if errors:
for error in errors:
print(error)
sys.exit(1)
else:
print("OK")
if __name__ == "__main__":
main()

View File

@@ -108,7 +108,7 @@ jobs:
runs-on: macos-latest
env:
MACOSX_DEPLOYMENT_TARGET: 10.9
MACOSX_DEPLOYMENT_TARGET: 10.14
steps:
- name: Checkout

View File

@@ -15,37 +15,7 @@ jobs:
fetch-depth: 4
- name: Get pull-request commits
run: |
set -x
# actions/checkout did a merge checkout of the pull-request. As such, the first
# commit is the merge commit. This means that on HEAD^ is the base branch, and
# on HEAD^2 are the commits from the pull-request. We now check if those trees
# have a common parent. If not, we fetch a few more commits till we do. In result,
# the log between HEAD^ and HEAD^2 will be the commits in the pull-request.
DEPTH=4
while [ -z "$(git merge-base HEAD^ HEAD^2)" ]; do
# Prevent infinite recursion
if [ ${DEPTH} -gt 256 ]; then
echo "No common parent between '${GITHUB_HEAD_REF}' and '${GITHUB_BASE_REF}'." >&2
exit 1
fi
git -c protocol.version=2 fetch --no-tags --prune --progress --no-recurse-submodules --deepen=${DEPTH} origin HEAD
DEPTH=$(( ${DEPTH} * 4 ))
done
# Just to show which commits we are going to evaluate.
git log --oneline HEAD^..HEAD^2
- name: Checkout commit-checker
uses: actions/checkout@v2
with:
repository: OpenTTD/OpenTTD-git-hooks
path: git-hooks
ref: master
uses: OpenTTD/actions/checkout-pull-request@v2
- name: Check commits
run: |
set -x
HOOKS_DIR=./git-hooks/hooks GIT_DIR=.git ./git-hooks/hooks/check-commits.sh HEAD^..HEAD^2
echo "Commit checks passed"
uses: OpenTTD/OpenTTD-git-hooks@main

View File

@@ -373,14 +373,19 @@ jobs:
include:
- container_image: "ubuntu:18.04"
bundle_name: "bionic"
compiler: "g++-8"
- container_image: "ubuntu:20.04"
bundle_name: "focal"
compiler: "g++"
- container_image: "ubuntu:20.10"
bundle_name: "groovy"
compiler: "g++"
- container_image: "debian:buster"
bundle_name: "buster"
compiler: "g++"
- container_image: "debian:bullseye"
bundle_name: "bullseye"
compiler: "g++"
runs-on: ubuntu-20.04
container:
@@ -406,7 +411,7 @@ jobs:
apt-get install -y --no-install-recommends \
cmake \
debhelper \
g++ \
${{ matrix.compiler }} \
git \
make \
openssl \
@@ -432,7 +437,7 @@ jobs:
cd build
echo "::group::CMake"
cmake ${GITHUB_WORKSPACE} \
CXX=${{ matrix.compiler }} cmake ${GITHUB_WORKSPACE} \
-DCMAKE_BUILD_TYPE=RelWithDebInfo \
-DCMAKE_INSTALL_PREFIX=/usr \
# EOF
@@ -470,7 +475,7 @@ jobs:
runs-on: macos-10.15
env:
MACOSX_DEPLOYMENT_TARGET: 10.9
MACOSX_DEPLOYMENT_TARGET: 10.14
steps:
- name: Download source
@@ -746,6 +751,21 @@ jobs:
with:
arch: ${{ matrix.host }}
- name: Import code signing certificate
shell: powershell
# If this is run on a fork, there may not be a certificate set up - continue in this case
continue-on-error: true
run: |
$tempFile = [System.IO.Path]::GetTempFileName()
$bytes = [System.Convert]::FromBase64String($env:WINDOWS_CERTIFICATE_P12)
[IO.File]::WriteAllBytes($tempFile, $bytes)
$pwd = ConvertTo-SecureString $env:WINDOWS_CERTIFICATE_PASSWORD -AsPlainText -Force
Import-PfxCertificate -FilePath $tempFile -CertStoreLocation Cert:\CurrentUser\My -Password $pwd
Remove-Item $tempFile
env:
WINDOWS_CERTIFICATE_P12: ${{ secrets.WINDOWS_CERTIFICATE_P12 }}
WINDOWS_CERTIFICATE_PASSWORD: ${{ secrets.WINDOWS_CERTIFICATE_PASSWORD }}
- name: Build (with installer)
if: needs.source.outputs.is_tag == 'true'
shell: bash
@@ -761,12 +781,15 @@ jobs:
-DOPTION_USE_NSIS=ON \
-DHOST_BINARY_DIR=${GITHUB_WORKSPACE}/build-host \
-DCMAKE_BUILD_TYPE=RelWithDebInfo \
-DWINDOWS_CERTIFICATE_COMMON_NAME="${WINDOWS_CERTIFICATE_COMMON_NAME}" \
# EOF
echo "::endgroup::"
echo "::group::Build"
cmake --build .
echo "::endgroup::"
env:
WINDOWS_CERTIFICATE_COMMON_NAME: ${{ secrets.WINDOWS_CERTIFICATE_COMMON_NAME }}
- name: Build (without installer)
if: needs.source.outputs.is_tag != 'true'
@@ -782,12 +805,15 @@ jobs:
-DCMAKE_TOOLCHAIN_FILE="c:\vcpkg\scripts\buildsystems\vcpkg.cmake" \
-DHOST_BINARY_DIR=${GITHUB_WORKSPACE}/build-host \
-DCMAKE_BUILD_TYPE=RelWithDebInfo \
-DWINDOWS_CERTIFICATE_COMMON_NAME="${WINDOWS_CERTIFICATE_COMMON_NAME}" \
# EOF
echo "::endgroup::"
echo "::group::Build"
cmake --build .
echo "::endgroup::"
env:
WINDOWS_CERTIFICATE_COMMON_NAME: ${{ secrets.WINDOWS_CERTIFICATE_COMMON_NAME }}
- name: Create bundles
shell: bash
@@ -809,6 +835,17 @@ jobs:
rm -f bundles/*.sha256
echo "::endgroup::"
- name: Sign installer
if: needs.source.outputs.is_tag == 'true'
shell: bash
# If this is run on a fork, there may not be a certificate set up - continue in this case
continue-on-error: true
run: |
cd ${GITHUB_WORKSPACE}/build/bundles
../../os/windows/sign.bat *.exe "${WINDOWS_CERTIFICATE_COMMON_NAME}"
env:
WINDOWS_CERTIFICATE_COMMON_NAME: ${{ secrets.WINDOWS_CERTIFICATE_COMMON_NAME }}
- name: Store bundles
uses: actions/upload-artifact@v2
with:

18
.github/workflows/unused-strings.yml vendored Normal file
View File

@@ -0,0 +1,18 @@
name: Unused strings
on:
pull_request:
jobs:
unused-strings:
name: Unused strings
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v2
- name: Check for unused strings
run: |
set -ex
python3 .github/unused-strings.py