appstream.py

# -*- Mode:Python; indent-tabs-mode:nil; tab-width:4 -*-
#
# Copyright 2017-2022 Canonical Ltd.
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 3 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that 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.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

"""Appstream metadata extractor."""

import contextlib
import operator
import os
from io import StringIO
from typing import List, Optional, cast

import lxml.etree
from xdg.DesktopEntry import DesktopEntry

from snapcraft import errors

from .extracted_metadata import ExtractedMetadata

_XSLT = """\
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
                xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="text"/>
<xsl:strip-space elements="*"/>

<xsl:template match="@* | node()">
    <xsl:copy>
        <xsl:apply-templates select="@* | node()[not(@xml:lang)] | node()[@xml:lang = en]" />
    </xsl:copy>
</xsl:template>

<xsl:template match="comment()"/>

<xsl:template match="p">
<xsl:text>&#xA;</xsl:text>
<xsl:apply-templates />
<xsl:text>&#xA;</xsl:text>
</xsl:template>

<xsl:template match="ul">
<xsl:text>&#xA;</xsl:text>
<xsl:for-each select="li[not(@xml:lang)] | li[@xml:lang = en]">
<xsl:text>- </xsl:text>
<xsl:apply-templates />
<xsl:text>&#xA;</xsl:text>
</xsl:for-each>
</xsl:template>

<xsl:template match="ol">
<xsl:text>&#xA;</xsl:text>
<xsl:for-each select="li[not(@xml:lang)] | li[@xml:lang = en]">
<xsl:number count="li[not(@xml:lang)] | li[@xml:lang = en]" format="1. "/>
<xsl:apply-templates />
<xsl:text>&#xA;</xsl:text>
</xsl:for-each>
</xsl:template>

<xsl:template match="em">
<xsl:text>_</xsl:text>
<xsl:apply-templates />
<xsl:text>_</xsl:text>
</xsl:template>

<xsl:template match="code">
<xsl:apply-templates />
</xsl:template>

</xsl:stylesheet>
"""


def extract(relpath: str, *, workdir: str) -> Optional[ExtractedMetadata]:
    """Extract appstream metadata.

    :param file_relpath: Relative path to the file containing metadata.
    :param workdir: The part working directory where the metadata file is located.

    :return: The extracted metadata, if any.
    """
    if not relpath.endswith(".metainfo.xml") and not relpath.endswith(".appdata.xml"):
        return None

    dom = _get_transformed_dom(os.path.join(workdir, relpath))

    common_id = _get_value_from_xml_element(dom, "id")
    summary = _get_value_from_xml_element(dom, "summary")
    description = _get_value_from_xml_element(dom, "description")
    title = _get_value_from_xml_element(dom, "name")
    version = _get_latest_release_from_nodes(dom.findall("releases/release"))

    desktop_file_paths = []
    desktop_file_ids = _get_desktop_file_ids_from_nodes(dom.findall("launchable"))
    # if there are no launchables, use the appstream id to take into
    # account the legacy appstream definitions
    if common_id and not desktop_file_ids:
        if common_id.endswith(".desktop"):
            desktop_file_ids.append(common_id)
        else:
            desktop_file_ids.append(common_id + ".desktop")

    for desktop_file_id in desktop_file_ids:
        desktop_file_path = _desktop_file_id_to_path(desktop_file_id, workdir=workdir)
        if desktop_file_path:
            desktop_file_paths.append(desktop_file_path)

    icon = _extract_icon(dom, workdir, desktop_file_paths)

    return ExtractedMetadata(
        common_id=common_id,
        title=title,
        summary=summary,
        description=description,
        version=version,
        icon=icon,
        desktop_file_paths=desktop_file_paths,
    )


def _get_transformed_dom(path: str):
    dom = _get_dom(path)
    transform = _get_xslt()
    return transform(dom)


def _get_dom(path: str) -> lxml.etree.ElementTree:
    try:
        return lxml.etree.parse(path)  # noqa S320
    except OSError as err:
        raise errors.SnapcraftError(str(err)) from err
    except lxml.etree.ParseError as err:
        raise errors.MetadataExtractionError(path, str(err)) from err


def _get_xslt():
    xslt = lxml.etree.parse(StringIO(_XSLT))  # noqa S320
    return lxml.etree.XSLT(xslt)


def _get_value_from_xml_element(tree, key) -> Optional[str]:
    node = tree.find(key)
    if node is not None and node.text:
        # Lines that should be empty end up with empty space after the
        # transformation. One example of this is seen for paragraphs (i.e.; <p>)
        # than hold list in then (i.e.; <ol> or <ul>) so we split all lines
        # here and strip any unwanted space.
        # TODO: Improve the XSLT to remove the need for this.
        return "\n".join([n.strip() for n in node.text.splitlines()]).strip()
    return None


def _get_latest_release_from_nodes(nodes) -> Optional[str]:
    for node in nodes:
        if "version" in node.attrib:
            return node.attrib["version"]
    return None


def _get_desktop_file_ids_from_nodes(nodes) -> List[str]:
    desktop_file_ids = []  # type: List[str]
    for node in nodes:
        if "type" in node.attrib and node.attrib["type"] == "desktop-id":
            desktop_file_ids.append(node.text.strip())
    return desktop_file_ids


def _desktop_file_id_to_path(desktop_file_id: str, *, workdir: str) -> Optional[str]:
    # For details about desktop file ids and their corresponding paths, see
    # https://standards.freedesktop.org/desktop-entry-spec/desktop-entry-spec-latest.html#desktop-file-id
    for xdg_data_dir in ("usr/local/share", "usr/share"):
        desktop_file_path = os.path.join(
            xdg_data_dir, "applications", desktop_file_id.replace("-", "/")
        )
        # Check if it exists in workdir, but do not add it to the resulting path
        # as it later needs to exist in the prime directory to effectively be
        # used.
        if os.path.exists(os.path.join(workdir, desktop_file_path)):
            return desktop_file_path
    return None


def _extract_icon(dom, workdir: str, desktop_file_paths: List[str]) -> Optional[str]:
    icon_node = dom.find("icon")
    if icon_node is not None and "type" in icon_node.attrib:
        icon_node_type = icon_node.attrib["type"]
    else:
        icon_node_type = None

    icon = icon_node.text.strip() if icon_node is not None else None

    if icon_node_type == "remote":
        return icon

    if icon_node_type == "stock" and icon is not None:
        return _get_icon_from_theme(workdir, "hicolor", icon)

    # If an icon path is specified and the icon file exists, we'll use that, otherwise
    # we'll fall back to what's listed in the desktop file.
    if icon is None:
        return _get_icon_from_desktop_file(workdir, desktop_file_paths)

    if os.path.exists(os.path.join(workdir, icon.lstrip("/"))):
        return icon

    return _get_icon_from_desktop_file(workdir, desktop_file_paths)


def _get_icon_from_desktop_file(
    workdir: str, desktop_file_paths: List[str]
) -> Optional[str]:
    # Icons in the desktop file can be either a full path to the icon file, or a name
    # to be searched in the standard locations. If the path is specified, use that,
    # otherwise look for the icon in the hicolor theme (also covers icon type="stock").
    # See https://standards.freedesktop.org/icon-theme-spec/icon-theme-spec-latest.html
    # for further information.
    for path in desktop_file_paths:
        entry = DesktopEntry()
        entry.parse(os.path.join(workdir, path))
        icon = cast(str, entry.getIcon())
        icon_path = (
            icon
            if os.path.isabs(icon)
            else _get_icon_from_theme(workdir, "hicolor", icon)
        )
        return icon_path

    return None


def _get_icon_from_theme(workdir: str, theme: str, icon: str) -> Optional[str]:
    # Icon themes can carry icons in different pre-rendered sizes or scalable. Scalable
    # implementation is optional, so we'll try the largest pixmap and then scalable if
    # no other sizes are available.
    theme_dir = os.path.join("usr", "share", "icons", theme)
    if not os.path.exists(os.path.join(workdir, theme_dir)):
        return None

    # TODO: use index.theme
    entries = os.listdir(os.path.join(workdir, theme_dir))
    # size is NxN
    x_entries = (e.split("x") for e in entries if "x" in e)
    sized_entries = (e[0] for e in x_entries if e[0] == e[1])
    sizes = {}
    for icon_size_entry in sized_entries:
        with contextlib.suppress(ValueError):
            isize = int(icon_size_entry)
            sizes[isize] = f"{isize}x{isize}"

    icon_size = None
    suffixes = []
    if sizes:
        size = max(sizes.items(), key=operator.itemgetter(1))[0]
        icon_size = sizes[size]
        suffixes = [".png", ".xpm"]
    elif "scalable" in entries:
        icon_size = "scalable"
        suffixes = [".svg", ".svgz"]

    icon_path = None
    if icon_size:
        for suffix in suffixes:
            icon_path = os.path.join(theme_dir, icon_size, "apps", icon + suffix)
            if os.path.exists(os.path.join(workdir, icon_path)):
                break

    return icon_path