wechat-article-parser 0.0.13__tar.gz → 0.0.15__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/PKG-INFO +31 -3
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/README.md +30 -2
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/pyproject.toml +1 -1
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/__init__.py +2 -1
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/models.py +21 -1
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/parser.py +97 -6
- wechat_article_parser-0.0.15/tests/test_article_urls.py +148 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_models.py +51 -1
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_paid_content.py +84 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_parser.py +4 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/.gitignore +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/LICENSE +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/conftest.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_markdown.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_miniprogram_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_product_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_profile_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_repost_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_svg_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_swiper_content.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_type_routing.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_user_name.py +0 -0
- {wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_video_content.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: wechat-article-parser
|
|
3
|
-
Version: 0.0.
|
|
3
|
+
Version: 0.0.15
|
|
4
4
|
Summary: WeChat MP article parser - extract metadata and content from WeChat public account articles
|
|
5
5
|
License-Expression: MIT
|
|
6
6
|
License-File: LICENSE
|
|
@@ -37,6 +37,7 @@ Markdown 转换前会展开无格式作用的 `span`,合并嵌套的重复加
|
|
|
37
37
|
图文正文仅包含视频号视频卡片或微信视频 iframe 时,Markdown 返回“这是一篇纯视频内容。”。
|
|
38
38
|
正文只有付费预览占位标签时,Markdown 返回“这是一篇付费内容,正文需付费后查看。”,
|
|
39
39
|
并保留公开元数据;已有的可见预览正文仍正常提取。
|
|
40
|
+
返回字段 `article_is_paid` 独立标记文章是否付费,有可见预览正文时也会识别。
|
|
40
41
|
图文中的公众号推荐卡片按原文顺序转换为 `**公众号名称**:简介`,每张卡片一段;
|
|
41
42
|
名称或简介缺失时,仅输出已有字段。
|
|
42
43
|
图文正文仅包含商品广告卡片(`mp-common-cpsad`)时,Markdown 返回“这是一篇商品推荐内容。”;
|
|
@@ -108,6 +109,17 @@ result = parse(
|
|
|
108
109
|
)
|
|
109
110
|
```
|
|
110
111
|
|
|
112
|
+
### 长链接
|
|
113
|
+
|
|
114
|
+
同步和异步接口均支持短链接 `/s/<22位文章ID>` 和长链接 `/s?__biz=...&mid=...&idx=...&sn=...&chksm=...`。
|
|
115
|
+
根据原始输入 URL 识别链接类型:协议须为 HTTP/HTTPS,域名须为 `mp.weixin.qq.com`。
|
|
116
|
+
长链接路径须为 `/s`,且 `__biz`、`mid`、`idx`、`sn`、`chksm` 五个查询参数必须都存在且有值,
|
|
117
|
+
空字符串或纯空白不算有效值;重复参数中的每个值也须非空白。额外查询参数和片段不影响识别。
|
|
118
|
+
未满足识别条件时,`article_url_type` 为 `unknown`,仍尝试解析页面,不额外抛出链接校验异常。
|
|
119
|
+
|
|
120
|
+
长链接返回的 `article_id` 为空,其他字段继续从页面提取。
|
|
121
|
+
只有识别为 `long` 时,`is_valid` 才允许缺少 `article_id`。
|
|
122
|
+
|
|
111
123
|
## 返回结构
|
|
112
124
|
|
|
113
125
|
`parse` 和 `parse_async` 均返回 `ArticleResult` 数据类,包含以下字段:
|
|
@@ -122,7 +134,8 @@ result = parse(
|
|
|
122
134
|
| `mp_image` | `str` | 公众号头像链接 |
|
|
123
135
|
| `mp_description` | `str` | 公众号简介 |
|
|
124
136
|
| `mp_account_type` | `AccountType` | 账号类型:`AccountType.SUBSCRIPTION`(订阅号)/ `AccountType.SERVICE`(服务号)/ `AccountType.UNKNOWN`(未识别) |
|
|
125
|
-
| `article_id` | `str` |
|
|
137
|
+
| `article_id` | `str` | 短链接中的文章 ID;长链接为空字符串 |
|
|
138
|
+
| `article_url_type` | `ArticleURLType` | 输入链接类型:`short`、`long` 或 `unknown`(默认);可从包顶层导入枚举,支持字符串比较和 JSON 序列化 |
|
|
126
139
|
| `article_msg_id` | `int` | 文章所在的群发消息 ID |
|
|
127
140
|
| `article_idx` | `int` | 群发图文中的位置(从 1 开始) |
|
|
128
141
|
| `article_sn` | `str` | 文章签名(防伪校验) |
|
|
@@ -131,16 +144,31 @@ result = parse(
|
|
|
131
144
|
| `article_description` | `str` | 文章摘要 |
|
|
132
145
|
| `article_markdown` | `str` | 文章正文的 Markdown 内容;视频或音频分享没有文字附言时可为空 |
|
|
133
146
|
| `article_publish_time` | `int` | 发布时间(Unix 时间戳) |
|
|
147
|
+
| `article_is_paid` | `bool` | 是否检测到当前文章的付费标记,默认 `False`;不表示是否已获取付费全文 |
|
|
134
148
|
| `article_display_type` | `ArticleDisplayType` | 文章展示类型:`article`、`video`、`audio`、`image`、`text` 或 `unknown` |
|
|
135
149
|
| `article_item_show_type` | `int \| None` | 微信页面中的原始展示类型:`0` 图文、`5` 视频、`7` 音频、`8` 图片轮播、`10` 文字;其他已提取的数值原样保留,缺失或无法解析时为 `None` |
|
|
136
150
|
| `raw_html` | `str` | 原始 HTML 网页源代码(仅在 `include_raw_html=True` 时填充,否则为空字符串) |
|
|
137
151
|
| `images` | `list[str]` | 文章中提取的所有图片链接 |
|
|
138
152
|
| `is_valid` | `bool` | 关键字段是否全部解析成功(属性) |
|
|
139
153
|
|
|
140
|
-
`is_valid` 为 `True` 的条件:`mp_id`、`mp_name`、`
|
|
154
|
+
`is_valid` 为 `True` 的条件:`mp_id`、`mp_name`、`article_msg_id`、`article_idx`、`article_sn`、`article_title`、`article_publish_time` 均不为空/零,且 `article_markdown` 非空或 `article_display_type in ("video", "audio")`。
|
|
155
|
+
此外,`article_url_type == "long"` 时允许 `article_id` 为空;短链接和未识别链接仍要求 `article_id` 非空。
|
|
156
|
+
视频和音频分享允许没有文字正文,仍须满足其他关键字段要求;有效性不表示已获取视频或音频文件。
|
|
141
157
|
|
|
142
158
|
付费受限页面返回提示且关键元数据完整时,`is_valid` 也为 `True`,这不表示已获取付费正文。
|
|
143
159
|
|
|
160
|
+
### 判断是否为付费文章
|
|
161
|
+
|
|
162
|
+
```python
|
|
163
|
+
result = parse("https://mp.weixin.qq.com/s/xxxxx")
|
|
164
|
+
print(result.article_is_paid) # True:识别为付费文章
|
|
165
|
+
```
|
|
166
|
+
|
|
167
|
+
`article_is_paid` 根据页面全局或 `window.cgiDataNew` 直接属性中的 `is_pay_subscribe`,
|
|
168
|
+
以及正文中的 `mp-pay-preview-filter` 标签识别;未检测到付费标记时返回 `False`。
|
|
169
|
+
该字段包含在 `dataclasses.asdict(result)` 中,JSON 序列化为布尔值。
|
|
170
|
+
有可见预览或可访问正文的付费文章仍返回 `True`,正文按原有逻辑提取。
|
|
171
|
+
|
|
144
172
|
### 判断文章展示类型
|
|
145
173
|
|
|
146
174
|
`ArticleDisplayType` 是字符串枚举,既支持枚举比较,也支持语义字符串比较:
|
|
@@ -22,6 +22,7 @@ Markdown 转换前会展开无格式作用的 `span`,合并嵌套的重复加
|
|
|
22
22
|
图文正文仅包含视频号视频卡片或微信视频 iframe 时,Markdown 返回“这是一篇纯视频内容。”。
|
|
23
23
|
正文只有付费预览占位标签时,Markdown 返回“这是一篇付费内容,正文需付费后查看。”,
|
|
24
24
|
并保留公开元数据;已有的可见预览正文仍正常提取。
|
|
25
|
+
返回字段 `article_is_paid` 独立标记文章是否付费,有可见预览正文时也会识别。
|
|
25
26
|
图文中的公众号推荐卡片按原文顺序转换为 `**公众号名称**:简介`,每张卡片一段;
|
|
26
27
|
名称或简介缺失时,仅输出已有字段。
|
|
27
28
|
图文正文仅包含商品广告卡片(`mp-common-cpsad`)时,Markdown 返回“这是一篇商品推荐内容。”;
|
|
@@ -93,6 +94,17 @@ result = parse(
|
|
|
93
94
|
)
|
|
94
95
|
```
|
|
95
96
|
|
|
97
|
+
### 长链接
|
|
98
|
+
|
|
99
|
+
同步和异步接口均支持短链接 `/s/<22位文章ID>` 和长链接 `/s?__biz=...&mid=...&idx=...&sn=...&chksm=...`。
|
|
100
|
+
根据原始输入 URL 识别链接类型:协议须为 HTTP/HTTPS,域名须为 `mp.weixin.qq.com`。
|
|
101
|
+
长链接路径须为 `/s`,且 `__biz`、`mid`、`idx`、`sn`、`chksm` 五个查询参数必须都存在且有值,
|
|
102
|
+
空字符串或纯空白不算有效值;重复参数中的每个值也须非空白。额外查询参数和片段不影响识别。
|
|
103
|
+
未满足识别条件时,`article_url_type` 为 `unknown`,仍尝试解析页面,不额外抛出链接校验异常。
|
|
104
|
+
|
|
105
|
+
长链接返回的 `article_id` 为空,其他字段继续从页面提取。
|
|
106
|
+
只有识别为 `long` 时,`is_valid` 才允许缺少 `article_id`。
|
|
107
|
+
|
|
96
108
|
## 返回结构
|
|
97
109
|
|
|
98
110
|
`parse` 和 `parse_async` 均返回 `ArticleResult` 数据类,包含以下字段:
|
|
@@ -107,7 +119,8 @@ result = parse(
|
|
|
107
119
|
| `mp_image` | `str` | 公众号头像链接 |
|
|
108
120
|
| `mp_description` | `str` | 公众号简介 |
|
|
109
121
|
| `mp_account_type` | `AccountType` | 账号类型:`AccountType.SUBSCRIPTION`(订阅号)/ `AccountType.SERVICE`(服务号)/ `AccountType.UNKNOWN`(未识别) |
|
|
110
|
-
| `article_id` | `str` |
|
|
122
|
+
| `article_id` | `str` | 短链接中的文章 ID;长链接为空字符串 |
|
|
123
|
+
| `article_url_type` | `ArticleURLType` | 输入链接类型:`short`、`long` 或 `unknown`(默认);可从包顶层导入枚举,支持字符串比较和 JSON 序列化 |
|
|
111
124
|
| `article_msg_id` | `int` | 文章所在的群发消息 ID |
|
|
112
125
|
| `article_idx` | `int` | 群发图文中的位置(从 1 开始) |
|
|
113
126
|
| `article_sn` | `str` | 文章签名(防伪校验) |
|
|
@@ -116,16 +129,31 @@ result = parse(
|
|
|
116
129
|
| `article_description` | `str` | 文章摘要 |
|
|
117
130
|
| `article_markdown` | `str` | 文章正文的 Markdown 内容;视频或音频分享没有文字附言时可为空 |
|
|
118
131
|
| `article_publish_time` | `int` | 发布时间(Unix 时间戳) |
|
|
132
|
+
| `article_is_paid` | `bool` | 是否检测到当前文章的付费标记,默认 `False`;不表示是否已获取付费全文 |
|
|
119
133
|
| `article_display_type` | `ArticleDisplayType` | 文章展示类型:`article`、`video`、`audio`、`image`、`text` 或 `unknown` |
|
|
120
134
|
| `article_item_show_type` | `int \| None` | 微信页面中的原始展示类型:`0` 图文、`5` 视频、`7` 音频、`8` 图片轮播、`10` 文字;其他已提取的数值原样保留,缺失或无法解析时为 `None` |
|
|
121
135
|
| `raw_html` | `str` | 原始 HTML 网页源代码(仅在 `include_raw_html=True` 时填充,否则为空字符串) |
|
|
122
136
|
| `images` | `list[str]` | 文章中提取的所有图片链接 |
|
|
123
137
|
| `is_valid` | `bool` | 关键字段是否全部解析成功(属性) |
|
|
124
138
|
|
|
125
|
-
`is_valid` 为 `True` 的条件:`mp_id`、`mp_name`、`
|
|
139
|
+
`is_valid` 为 `True` 的条件:`mp_id`、`mp_name`、`article_msg_id`、`article_idx`、`article_sn`、`article_title`、`article_publish_time` 均不为空/零,且 `article_markdown` 非空或 `article_display_type in ("video", "audio")`。
|
|
140
|
+
此外,`article_url_type == "long"` 时允许 `article_id` 为空;短链接和未识别链接仍要求 `article_id` 非空。
|
|
141
|
+
视频和音频分享允许没有文字正文,仍须满足其他关键字段要求;有效性不表示已获取视频或音频文件。
|
|
126
142
|
|
|
127
143
|
付费受限页面返回提示且关键元数据完整时,`is_valid` 也为 `True`,这不表示已获取付费正文。
|
|
128
144
|
|
|
145
|
+
### 判断是否为付费文章
|
|
146
|
+
|
|
147
|
+
```python
|
|
148
|
+
result = parse("https://mp.weixin.qq.com/s/xxxxx")
|
|
149
|
+
print(result.article_is_paid) # True:识别为付费文章
|
|
150
|
+
```
|
|
151
|
+
|
|
152
|
+
`article_is_paid` 根据页面全局或 `window.cgiDataNew` 直接属性中的 `is_pay_subscribe`,
|
|
153
|
+
以及正文中的 `mp-pay-preview-filter` 标签识别;未检测到付费标记时返回 `False`。
|
|
154
|
+
该字段包含在 `dataclasses.asdict(result)` 中,JSON 序列化为布尔值。
|
|
155
|
+
有可见预览或可访问正文的付费文章仍返回 `True`,正文按原有逻辑提取。
|
|
156
|
+
|
|
129
157
|
### 判断文章展示类型
|
|
130
158
|
|
|
131
159
|
`ArticleDisplayType` 是字符串枚举,既支持枚举比较,也支持语义字符串比较:
|
|
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
|
|
|
4
4
|
|
|
5
5
|
[project]
|
|
6
6
|
name = "wechat-article-parser"
|
|
7
|
-
version = "0.0.
|
|
7
|
+
version = "0.0.15"
|
|
8
8
|
description = "WeChat MP article parser - extract metadata and content from WeChat public account articles"
|
|
9
9
|
readme = "README.md"
|
|
10
10
|
requires-python = ">=3.10"
|
{wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/__init__.py
RENAMED
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
from .models import AccountType, ArticleDisplayType, ArticleResult, WeChatVerifyError
|
|
1
|
+
from .models import AccountType, ArticleDisplayType, ArticleResult, ArticleURLType, WeChatVerifyError
|
|
2
2
|
from .parser import parse, parse_async
|
|
3
3
|
|
|
4
4
|
__all__ = [
|
|
@@ -7,5 +7,6 @@ __all__ = [
|
|
|
7
7
|
"ArticleResult",
|
|
8
8
|
"AccountType",
|
|
9
9
|
"ArticleDisplayType",
|
|
10
|
+
"ArticleURLType",
|
|
10
11
|
"WeChatVerifyError",
|
|
11
12
|
]
|
{wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/models.py
RENAMED
|
@@ -22,6 +22,20 @@ class AccountType(str, Enum):
|
|
|
22
22
|
return format(self.value, format_spec)
|
|
23
23
|
|
|
24
24
|
|
|
25
|
+
class ArticleURLType(str, Enum):
|
|
26
|
+
"""输入链接类型;仅通过长链接校验时才允许缺少文章 ID。"""
|
|
27
|
+
|
|
28
|
+
UNKNOWN = "unknown"
|
|
29
|
+
SHORT = "short"
|
|
30
|
+
LONG = "long"
|
|
31
|
+
|
|
32
|
+
def __str__(self) -> str:
|
|
33
|
+
return self.value
|
|
34
|
+
|
|
35
|
+
def __format__(self, format_spec: str) -> str:
|
|
36
|
+
return format(self.value, format_spec)
|
|
37
|
+
|
|
38
|
+
|
|
25
39
|
class ArticleDisplayType(str, Enum):
|
|
26
40
|
"""文章展示类型,支持枚举和语义字符串比较。"""
|
|
27
41
|
|
|
@@ -76,13 +90,19 @@ class ArticleResult:
|
|
|
76
90
|
# 当前公众号的页面全局 user_name,未提取到时为空字符串。
|
|
77
91
|
mp_user_name: str = ""
|
|
78
92
|
|
|
93
|
+
# 页面是否包含当前文章的付费标记;不表示是否已获取付费全文。
|
|
94
|
+
article_is_paid: bool = False
|
|
95
|
+
|
|
96
|
+
# 根据原始输入 URL 识别,不根据 article_id 是否为空反推。
|
|
97
|
+
article_url_type: ArticleURLType = ArticleURLType.UNKNOWN
|
|
98
|
+
|
|
79
99
|
@property
|
|
80
100
|
def is_valid(self) -> bool:
|
|
81
101
|
"""检查关键字段是否解析成功;视频和音频分享允许正文为空。"""
|
|
82
102
|
return bool(
|
|
83
103
|
self.mp_id
|
|
84
104
|
and self.mp_name
|
|
85
|
-
and self.article_id
|
|
105
|
+
and (self.article_id or self.article_url_type == ArticleURLType.LONG)
|
|
86
106
|
and self.article_msg_id
|
|
87
107
|
and self.article_idx
|
|
88
108
|
and self.article_sn
|
{wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/src/wechat_article_parser/parser.py
RENAMED
|
@@ -7,13 +7,13 @@ import html as html_module
|
|
|
7
7
|
import re
|
|
8
8
|
from collections import deque
|
|
9
9
|
from dataclasses import replace
|
|
10
|
-
from urllib.parse import unquote
|
|
10
|
+
from urllib.parse import parse_qs, unquote, urlsplit
|
|
11
11
|
|
|
12
12
|
import httpx
|
|
13
13
|
from bs4 import BeautifulSoup, Tag
|
|
14
14
|
from markdownify import MarkdownConverter
|
|
15
15
|
|
|
16
|
-
from .models import AccountType, ArticleDisplayType, ArticleResult, WeChatVerifyError
|
|
16
|
+
from .models import AccountType, ArticleDisplayType, ArticleResult, ArticleURLType, WeChatVerifyError
|
|
17
17
|
|
|
18
18
|
_USER_AGENT = (
|
|
19
19
|
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) "
|
|
@@ -216,10 +216,33 @@ def _extract_picture_cdn_urls(script_text: str) -> list[str]:
|
|
|
216
216
|
# 内部提取函数
|
|
217
217
|
# ---------------------------------------------------------------------------
|
|
218
218
|
|
|
219
|
+
def _extract_article_url_type(url: str) -> ArticleURLType:
|
|
220
|
+
"""识别输入链接;长链接的五个必需参数均须存在且非空白。"""
|
|
221
|
+
try:
|
|
222
|
+
parts = urlsplit(url)
|
|
223
|
+
except ValueError:
|
|
224
|
+
return ArticleURLType.UNKNOWN
|
|
225
|
+
if (
|
|
226
|
+
parts.scheme not in ("http", "https")
|
|
227
|
+
or parts.hostname != "mp.weixin.qq.com"
|
|
228
|
+
or parts.username is not None
|
|
229
|
+
or parts.password is not None
|
|
230
|
+
):
|
|
231
|
+
return ArticleURLType.UNKNOWN
|
|
232
|
+
if re.fullmatch(r"/s/[A-Za-z0-9_-]{22}", parts.path):
|
|
233
|
+
return ArticleURLType.SHORT
|
|
234
|
+
if parts.path == "/s":
|
|
235
|
+
params = parse_qs(parts.query, keep_blank_values=True)
|
|
236
|
+
required = ("__biz", "mid", "idx", "sn", "chksm")
|
|
237
|
+
# 保留空值,重复参数中出现空值也不能放行。
|
|
238
|
+
if all(key in params and all(value.strip() for value in params[key]) for key in required):
|
|
239
|
+
return ArticleURLType.LONG
|
|
240
|
+
return ArticleURLType.UNKNOWN
|
|
241
|
+
|
|
242
|
+
|
|
219
243
|
def _extract_article_id(url: str) -> str:
|
|
220
|
-
|
|
221
|
-
|
|
222
|
-
return parts[4]
|
|
244
|
+
if _extract_article_url_type(url) == ArticleURLType.SHORT:
|
|
245
|
+
return urlsplit(url).path.rsplit("/", 1)[1]
|
|
223
246
|
return ""
|
|
224
247
|
|
|
225
248
|
|
|
@@ -272,6 +295,69 @@ def _extract_item_show_type(scripts: list[Tag]) -> int | None:
|
|
|
272
295
|
return None
|
|
273
296
|
|
|
274
297
|
|
|
298
|
+
def _extract_is_paid(soup: BeautifulSoup, scripts: list[Tag]) -> bool:
|
|
299
|
+
"""识别当前文章的付费标记,排除正文外占位符和嵌套推荐文章的数据。"""
|
|
300
|
+
if soup.select_one(
|
|
301
|
+
"div.rich_media_content mp-pay-preview-filter, "
|
|
302
|
+
"div.original_page mp-pay-preview-filter"
|
|
303
|
+
) is not None:
|
|
304
|
+
return True
|
|
305
|
+
|
|
306
|
+
for script in scripts:
|
|
307
|
+
if "is_pay_subscribe" not in script.text:
|
|
308
|
+
continue
|
|
309
|
+
tokens = [
|
|
310
|
+
match.group() for match in _JS_DATA_TOKEN_RE.finditer(script.text)
|
|
311
|
+
if match.lastgroup != "comment"
|
|
312
|
+
]
|
|
313
|
+
stack: list[str] = []
|
|
314
|
+
cgi_depth: int | None = None
|
|
315
|
+
for i, token in enumerate(tokens):
|
|
316
|
+
if token == "{":
|
|
317
|
+
if (
|
|
318
|
+
tokens[max(0, i - 4):i] == ["window", ".", "cgiDataNew", "="]
|
|
319
|
+
and (i < 5 or tokens[i - 5] != ".")
|
|
320
|
+
):
|
|
321
|
+
cgi_depth = len(stack) + 1
|
|
322
|
+
stack.append(token)
|
|
323
|
+
elif token in ("[", "("):
|
|
324
|
+
stack.append(token)
|
|
325
|
+
elif token in ("}", "]", ")"):
|
|
326
|
+
if stack and stack[-1] == {"}": "{", "]": "[", ")": "("}[token]:
|
|
327
|
+
if len(stack) == cgi_depth:
|
|
328
|
+
cgi_depth = None
|
|
329
|
+
stack.pop()
|
|
330
|
+
if token.strip("\"'") != "is_pay_subscribe":
|
|
331
|
+
continue
|
|
332
|
+
is_global = token == "is_pay_subscribe" and (
|
|
333
|
+
(not stack and i > 0 and tokens[i - 1] in ("var", "let", "const"))
|
|
334
|
+
or (
|
|
335
|
+
tokens[max(0, i - 2):i] == ["window", "."]
|
|
336
|
+
and (i < 3 or tokens[i - 3] != ".")
|
|
337
|
+
)
|
|
338
|
+
)
|
|
339
|
+
is_cgi_field = (
|
|
340
|
+
cgi_depth is not None and len(stack) == cgi_depth
|
|
341
|
+
and i > 0 and tokens[i - 1] in ("{", ",")
|
|
342
|
+
)
|
|
343
|
+
separator = tokens[i + 1:i + 2]
|
|
344
|
+
if not (
|
|
345
|
+
(is_global and separator == ["="])
|
|
346
|
+
or (is_cgi_field and separator == [":"])
|
|
347
|
+
):
|
|
348
|
+
continue
|
|
349
|
+
# 仅识别页面使用的字面量及乘 1 转数值写法,不执行 JS 表达式。
|
|
350
|
+
value = tokens[i + 2:i + 3]
|
|
351
|
+
if value not in (["1"], ["'1'"], ['"1"'], ["true"]):
|
|
352
|
+
continue
|
|
353
|
+
end = i + 3
|
|
354
|
+
if value != ["true"] and tokens[end:end + 2] == ["*", "1"]:
|
|
355
|
+
end += 2
|
|
356
|
+
if end == len(tokens) or tokens[end] in (",", ";", "}"):
|
|
357
|
+
return True
|
|
358
|
+
return False
|
|
359
|
+
|
|
360
|
+
|
|
275
361
|
def _extract_user_name(scripts: list[Tag]) -> str:
|
|
276
362
|
"""读取全局 user_name,缺失时读取 window.cgiDataNew 的直接属性。
|
|
277
363
|
|
|
@@ -469,6 +555,8 @@ def _extract_rich_media_content(content_tag: Tag, result: ArticleResult) -> None
|
|
|
469
555
|
"""将 rich_media_content 区块转换为 Markdown。"""
|
|
470
556
|
# prettify 会在行内标签之间插入空白,改变代码和引用链接的内容。
|
|
471
557
|
soup = BeautifulSoup(str(content_tag), "html.parser")
|
|
558
|
+
if soup.find("mp-pay-preview-filter") is not None:
|
|
559
|
+
result.article_is_paid = True
|
|
472
560
|
has_video = soup.select_one(
|
|
473
561
|
'mp-common-videosnap[data-type="video"], iframe.video_iframe, iframe.wx_video_iframe'
|
|
474
562
|
) is not None
|
|
@@ -774,12 +862,15 @@ def _parse_html(url: str, html: str) -> ArticleResult:
|
|
|
774
862
|
result = ArticleResult()
|
|
775
863
|
soup = BeautifulSoup(html, "html.parser")
|
|
776
864
|
|
|
865
|
+
result.article_url_type = _extract_article_url_type(url)
|
|
777
866
|
result.article_id = _extract_article_id(url)
|
|
778
867
|
_extract_meta(soup, result)
|
|
779
868
|
|
|
780
869
|
scripts = soup.find_all("script", attrs={"type": "text/javascript"})
|
|
781
870
|
|
|
782
|
-
|
|
871
|
+
all_scripts = soup.find_all("script")
|
|
872
|
+
result.mp_user_name = _extract_user_name(all_scripts)
|
|
873
|
+
result.article_is_paid = _extract_is_paid(soup, all_scripts)
|
|
783
874
|
item_show_type = _extract_item_show_type(scripts)
|
|
784
875
|
result.article_item_show_type = item_show_type
|
|
785
876
|
result.article_display_type = _ARTICLE_DISPLAY_TYPES.get(
|
|
@@ -0,0 +1,148 @@
|
|
|
1
|
+
"""长短链接识别及有效性判断的离线回归测试。"""
|
|
2
|
+
|
|
3
|
+
from dataclasses import replace
|
|
4
|
+
from urllib.parse import urlencode
|
|
5
|
+
|
|
6
|
+
import httpx
|
|
7
|
+
import pytest
|
|
8
|
+
|
|
9
|
+
from wechat_article_parser import ArticleURLType, parser
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
_SHORT_URL = "https://mp.weixin.qq.com/s/ncPoZnxKdvDRLM_s4DBgmA"
|
|
13
|
+
_PARAMS = {
|
|
14
|
+
"__biz": "MjY2NzgwMjU0MA==",
|
|
15
|
+
"mid": "2650331950",
|
|
16
|
+
"idx": "3",
|
|
17
|
+
"sn": "59630a9449c484f555c71978ee521cd8",
|
|
18
|
+
"chksm": "ae4b5d1c0343c5bf3af413da074abb54a3d56268937ba1825fa518c6ba4b1c96c34675d15d9e",
|
|
19
|
+
}
|
|
20
|
+
_LONG_URL = "https://mp.weixin.qq.com/s?" + urlencode(_PARAMS)
|
|
21
|
+
_HTML = """
|
|
22
|
+
<meta property="og:title" content="长链接测试文章">
|
|
23
|
+
<div class="rich_media_content"><p>正文</p></div>
|
|
24
|
+
<script type="text/javascript">
|
|
25
|
+
var item_show_type = "0";
|
|
26
|
+
var hd_head_img = "https://example.com/avatar";
|
|
27
|
+
var nickname = htmlDecode("测试公众号");
|
|
28
|
+
var oriCreateTime = '1791884914';
|
|
29
|
+
window.__allowLoadResFromMp = true;
|
|
30
|
+
var biz = "MjY2NzgwMjU0MA==";
|
|
31
|
+
var mid = "2650331950";
|
|
32
|
+
var idx = "3";
|
|
33
|
+
var sn = "59630a9449c484f555c71978ee521cd8";
|
|
34
|
+
</script>
|
|
35
|
+
"""
|
|
36
|
+
|
|
37
|
+
|
|
38
|
+
@pytest.mark.parametrize("scheme", ["http", "https"])
|
|
39
|
+
@pytest.mark.parametrize("suffix", ["", "&scene=21#wechat_redirect"])
|
|
40
|
+
def test_long_url_parses_without_article_id(scheme: str, suffix: str) -> None:
|
|
41
|
+
result = parser._parse_html(_LONG_URL.replace("https:", scheme + ":") + suffix, _HTML)
|
|
42
|
+
|
|
43
|
+
assert result.article_url_type is ArticleURLType.LONG
|
|
44
|
+
assert result.article_id == ""
|
|
45
|
+
assert result.mp_id == 2667802540
|
|
46
|
+
assert result.article_msg_id == 2650331950
|
|
47
|
+
assert result.article_idx == 3
|
|
48
|
+
assert result.article_sn == _PARAMS["sn"]
|
|
49
|
+
assert result.article_markdown == "正文"
|
|
50
|
+
assert result.is_valid
|
|
51
|
+
|
|
52
|
+
|
|
53
|
+
@pytest.mark.parametrize("key", _PARAMS)
|
|
54
|
+
@pytest.mark.parametrize("value", [None, "", " \t\n"])
|
|
55
|
+
def test_long_url_requires_each_parameter_with_value(key: str, value: str | None) -> None:
|
|
56
|
+
params = _PARAMS.copy()
|
|
57
|
+
if value is None:
|
|
58
|
+
del params[key]
|
|
59
|
+
else:
|
|
60
|
+
params[key] = value
|
|
61
|
+
result = parser._parse_html("https://mp.weixin.qq.com/s?" + urlencode(params), _HTML)
|
|
62
|
+
|
|
63
|
+
assert result.article_url_type is ArticleURLType.UNKNOWN
|
|
64
|
+
assert result.article_id == ""
|
|
65
|
+
assert not result.is_valid
|
|
66
|
+
|
|
67
|
+
|
|
68
|
+
@pytest.mark.parametrize("key", _PARAMS)
|
|
69
|
+
def test_duplicate_blank_parameter_does_not_pass_validation(key: str) -> None:
|
|
70
|
+
result = parser._parse_html(_LONG_URL + "&" + key + "=", _HTML)
|
|
71
|
+
assert result.article_url_type is ArticleURLType.UNKNOWN
|
|
72
|
+
assert not result.is_valid
|
|
73
|
+
|
|
74
|
+
|
|
75
|
+
@pytest.mark.parametrize(
|
|
76
|
+
"base",
|
|
77
|
+
[
|
|
78
|
+
"https://example.com/s",
|
|
79
|
+
"https://mp.weixin.qq.com.example.com/s",
|
|
80
|
+
"https://mp.weixin.qq.com@evil.example/s",
|
|
81
|
+
"https://user@mp.weixin.qq.com/s",
|
|
82
|
+
"ftp://mp.weixin.qq.com/s",
|
|
83
|
+
"//mp.weixin.qq.com/s",
|
|
84
|
+
"https://mp.weixin.qq.com/s/",
|
|
85
|
+
"https://mp.weixin.qq.com/other",
|
|
86
|
+
"https://[invalid/s",
|
|
87
|
+
],
|
|
88
|
+
)
|
|
89
|
+
def test_unrecognized_url_cannot_skip_article_id(base: str) -> None:
|
|
90
|
+
result = parser._parse_html(base + "?" + urlencode(_PARAMS), _HTML)
|
|
91
|
+
assert result.article_url_type is ArticleURLType.UNKNOWN
|
|
92
|
+
assert not result.is_valid
|
|
93
|
+
|
|
94
|
+
|
|
95
|
+
def test_required_parameters_in_fragment_do_not_count() -> None:
|
|
96
|
+
result = parser._parse_html("https://mp.weixin.qq.com/s#" + urlencode(_PARAMS), _HTML)
|
|
97
|
+
assert result.article_url_type is ArticleURLType.UNKNOWN
|
|
98
|
+
assert not result.is_valid
|
|
99
|
+
|
|
100
|
+
|
|
101
|
+
@pytest.mark.parametrize("suffix", ["", "?scene=21#wechat_redirect", "?" + urlencode(_PARAMS)])
|
|
102
|
+
def test_short_url_still_requires_article_id(suffix: str) -> None:
|
|
103
|
+
result = parser._parse_html(_SHORT_URL + suffix, _HTML)
|
|
104
|
+
assert result.article_url_type is ArticleURLType.SHORT
|
|
105
|
+
assert result.article_id == "ncPoZnxKdvDRLM_s4DBgmA"
|
|
106
|
+
assert result.is_valid
|
|
107
|
+
assert not replace(result, article_id="").is_valid
|
|
108
|
+
|
|
109
|
+
|
|
110
|
+
def test_query_parameters_do_not_replace_missing_page_metadata() -> None:
|
|
111
|
+
result = parser._parse_html(_LONG_URL, '<div class="rich_media_content">正文</div>')
|
|
112
|
+
assert result.article_url_type is ArticleURLType.LONG
|
|
113
|
+
assert not result.is_valid
|
|
114
|
+
|
|
115
|
+
|
|
116
|
+
@pytest.fixture
|
|
117
|
+
def mock_article_http(monkeypatch):
|
|
118
|
+
# 重定向到短链接后仍根据原始输入链接识别类型。
|
|
119
|
+
def respond(request):
|
|
120
|
+
if request.url.path == "/s":
|
|
121
|
+
return httpx.Response(302, headers={"Location": _SHORT_URL})
|
|
122
|
+
return httpx.Response(200, text=_HTML)
|
|
123
|
+
|
|
124
|
+
transport = httpx.MockTransport(respond)
|
|
125
|
+
client, async_client = httpx.Client, httpx.AsyncClient
|
|
126
|
+
monkeypatch.setattr(
|
|
127
|
+
parser.httpx, "Client", lambda **kwargs: client(transport=transport, **kwargs),
|
|
128
|
+
)
|
|
129
|
+
monkeypatch.setattr(
|
|
130
|
+
parser.httpx, "AsyncClient", lambda **kwargs: async_client(transport=transport, **kwargs),
|
|
131
|
+
)
|
|
132
|
+
|
|
133
|
+
|
|
134
|
+
def test_sync_public_entry_accepts_long_url(mock_article_http) -> None:
|
|
135
|
+
result = parser.parse(_LONG_URL, include_raw_html=True)
|
|
136
|
+
assert result.article_url_type is ArticleURLType.LONG
|
|
137
|
+
assert result.article_id == ""
|
|
138
|
+
assert result.is_valid
|
|
139
|
+
assert result.raw_html == _HTML
|
|
140
|
+
|
|
141
|
+
|
|
142
|
+
@pytest.mark.asyncio
|
|
143
|
+
async def test_async_public_entry_accepts_long_url(mock_article_http) -> None:
|
|
144
|
+
result = await parser.parse_async(_LONG_URL, include_raw_html=True)
|
|
145
|
+
assert result.article_url_type is ArticleURLType.LONG
|
|
146
|
+
assert result.article_id == ""
|
|
147
|
+
assert result.is_valid
|
|
148
|
+
assert result.raw_html == _HTML
|
|
@@ -5,7 +5,7 @@ import json
|
|
|
5
5
|
|
|
6
6
|
import pytest
|
|
7
7
|
|
|
8
|
-
from wechat_article_parser import ArticleDisplayType, ArticleResult
|
|
8
|
+
from wechat_article_parser import ArticleDisplayType, ArticleResult, ArticleURLType
|
|
9
9
|
|
|
10
10
|
|
|
11
11
|
_COMPLETE = ArticleResult(
|
|
@@ -91,3 +91,53 @@ def test_default_display_type_is_unknown() -> None:
|
|
|
91
91
|
assert result.article_display_type is ArticleDisplayType.UNKNOWN
|
|
92
92
|
assert result.article_item_show_type is None
|
|
93
93
|
assert not result.is_valid
|
|
94
|
+
|
|
95
|
+
|
|
96
|
+
@pytest.mark.parametrize("is_paid", [False, True])
|
|
97
|
+
def test_paid_field_serializes_as_boolean(is_paid: bool) -> None:
|
|
98
|
+
result = ArticleResult(article_is_paid=is_paid)
|
|
99
|
+
assert json.loads(json.dumps(asdict(result)))["article_is_paid"] is is_paid
|
|
100
|
+
assert ArticleResult().article_is_paid is False
|
|
101
|
+
|
|
102
|
+
|
|
103
|
+
@pytest.mark.parametrize("url_type", list(ArticleURLType))
|
|
104
|
+
@pytest.mark.parametrize("display_type", list(ArticleDisplayType))
|
|
105
|
+
@pytest.mark.parametrize("markdown", ["", "正文"])
|
|
106
|
+
def test_only_long_url_can_omit_article_id(
|
|
107
|
+
url_type: ArticleURLType, display_type: ArticleDisplayType, markdown: str,
|
|
108
|
+
) -> None:
|
|
109
|
+
result = replace(
|
|
110
|
+
_COMPLETE, article_id="", article_url_type=url_type,
|
|
111
|
+
article_display_type=display_type, article_markdown=markdown,
|
|
112
|
+
)
|
|
113
|
+
assert result.is_valid is (
|
|
114
|
+
url_type == ArticleURLType.LONG
|
|
115
|
+
and (bool(markdown) or display_type in (ArticleDisplayType.VIDEO, ArticleDisplayType.AUDIO))
|
|
116
|
+
)
|
|
117
|
+
|
|
118
|
+
|
|
119
|
+
@pytest.mark.parametrize(
|
|
120
|
+
("field_name", "empty_value"),
|
|
121
|
+
[
|
|
122
|
+
("mp_id", 0), ("mp_name", ""), ("article_msg_id", 0),
|
|
123
|
+
("article_idx", 0), ("article_sn", ""), ("article_title", ""),
|
|
124
|
+
("article_publish_time", 0),
|
|
125
|
+
],
|
|
126
|
+
)
|
|
127
|
+
def test_long_url_still_requires_other_key_metadata(
|
|
128
|
+
field_name: str, empty_value: str | int,
|
|
129
|
+
) -> None:
|
|
130
|
+
result = replace(
|
|
131
|
+
_COMPLETE, article_id="", article_url_type=ArticleURLType.LONG,
|
|
132
|
+
article_markdown="正文", **{field_name: empty_value},
|
|
133
|
+
)
|
|
134
|
+
assert not result.is_valid
|
|
135
|
+
|
|
136
|
+
|
|
137
|
+
@pytest.mark.parametrize("url_type", list(ArticleURLType))
|
|
138
|
+
def test_url_type_serializes_as_semantic_string(url_type: ArticleURLType) -> None:
|
|
139
|
+
result = ArticleResult(article_url_type=url_type)
|
|
140
|
+
assert str(url_type) == url_type.value
|
|
141
|
+
assert f"{url_type:>10}" == f"{url_type.value:>10}"
|
|
142
|
+
assert json.loads(json.dumps(asdict(result)))["article_url_type"] == url_type.value
|
|
143
|
+
assert ArticleResult().article_url_type is ArticleURLType.UNKNOWN
|
|
@@ -51,6 +51,7 @@ def test_paid_placeholder_keeps_metadata_and_does_not_use_unrelated_images(
|
|
|
51
51
|
result = _parse_html(_URL, _html(_FILTER, metadata, data))
|
|
52
52
|
|
|
53
53
|
assert result.article_markdown == _NOTICE
|
|
54
|
+
assert result.article_is_paid is True
|
|
54
55
|
assert result.images == []
|
|
55
56
|
assert result.article_display_type is ArticleDisplayType.ARTICLE
|
|
56
57
|
assert result.article_item_show_type == 0
|
|
@@ -73,6 +74,7 @@ def test_paid_filter_does_not_replace_visible_preview(body: str) -> None:
|
|
|
73
74
|
result = _parse_html(_URL, _html(body + _FILTER))
|
|
74
75
|
|
|
75
76
|
assert result.is_valid
|
|
77
|
+
assert result.article_is_paid is True
|
|
76
78
|
assert _NOTICE not in result.article_markdown
|
|
77
79
|
if "免费预览" in body:
|
|
78
80
|
assert result.article_markdown == "可见的免费预览正文。"
|
|
@@ -86,6 +88,7 @@ def test_paid_filter_takes_priority_over_empty_video_embed() -> None:
|
|
|
86
88
|
result = _parse_html(_URL, _html(body))
|
|
87
89
|
|
|
88
90
|
assert result.is_valid
|
|
91
|
+
assert result.article_is_paid is True
|
|
89
92
|
assert result.article_markdown == _NOTICE
|
|
90
93
|
|
|
91
94
|
|
|
@@ -93,6 +96,7 @@ def test_paid_marker_outside_article_does_not_fill_empty_body() -> None:
|
|
|
93
96
|
result = _parse_html(_URL, _html("") + _FILTER)
|
|
94
97
|
|
|
95
98
|
assert result.article_markdown == ""
|
|
99
|
+
assert result.article_is_paid is False
|
|
96
100
|
assert not result.is_valid
|
|
97
101
|
|
|
98
102
|
|
|
@@ -103,4 +107,84 @@ def test_paid_flag_without_filter_keeps_accessible_body() -> None:
|
|
|
103
107
|
)
|
|
104
108
|
|
|
105
109
|
assert result.is_valid
|
|
110
|
+
assert result.article_is_paid is True
|
|
106
111
|
assert result.article_markdown == "已提供的完整正文。"
|
|
112
|
+
|
|
113
|
+
|
|
114
|
+
@pytest.mark.parametrize("value", ["1", "'1'", '"1"', "'1' * 1", '"1" * 1', "true"])
|
|
115
|
+
@pytest.mark.parametrize("template", [
|
|
116
|
+
"var is_pay_subscribe = VALUE;",
|
|
117
|
+
"window.is_pay_subscribe = VALUE;",
|
|
118
|
+
"window.cgiDataNew = {is_pay_subscribe: VALUE};",
|
|
119
|
+
'window.cgiDataNew = {"is_pay_subscribe": VALUE, other: 0};',
|
|
120
|
+
])
|
|
121
|
+
def test_paid_flag_forms(value: str, template: str) -> None:
|
|
122
|
+
result = _parse_html(
|
|
123
|
+
_URL, _html("<p>可见正文</p>", data=template.replace("VALUE", value)),
|
|
124
|
+
)
|
|
125
|
+
|
|
126
|
+
assert result.article_is_paid is True
|
|
127
|
+
assert result.article_markdown == "可见正文"
|
|
128
|
+
assert result.is_valid
|
|
129
|
+
|
|
130
|
+
|
|
131
|
+
@pytest.mark.parametrize("data", [
|
|
132
|
+
"",
|
|
133
|
+
"var is_pay_subscribe = 0;",
|
|
134
|
+
"window.cgiDataNew = {is_pay_subscribe: '0' * 1};",
|
|
135
|
+
"window.cgiDataNew = {is_pay_subscribe: false};",
|
|
136
|
+
"window.cgiDataNew = {is_pay_subscribe: '10'};",
|
|
137
|
+
"var is_pay_subscribe = '1' * 0;",
|
|
138
|
+
"var is_pay_subscribe = 1 === 0;",
|
|
139
|
+
"// var is_pay_subscribe = 1;",
|
|
140
|
+
"/* window.cgiDataNew = {is_pay_subscribe: 1}; */",
|
|
141
|
+
'var text = "var is_pay_subscribe = 1;";',
|
|
142
|
+
"var recommendation = {is_pay_subscribe: 1};",
|
|
143
|
+
"window.cgiDataNew = {card: {is_pay_subscribe: 1}, is_pay_subscribe: 0};",
|
|
144
|
+
"window.cgiDataNew = {cards: [{is_pay_subscribe: 1}]};",
|
|
145
|
+
"function unrelated() { var is_pay_subscribe = 1; }",
|
|
146
|
+
"other.window.cgiDataNew = {is_pay_subscribe: 1};",
|
|
147
|
+
"other.window.is_pay_subscribe = 1;",
|
|
148
|
+
])
|
|
149
|
+
def test_missing_free_or_unrelated_paid_flags(data: str) -> None:
|
|
150
|
+
result = _parse_html(_URL, _html("<p>正文提到付费,并不代表本文收费。</p>", data=data))
|
|
151
|
+
|
|
152
|
+
assert result.article_is_paid is False
|
|
153
|
+
assert result.is_valid
|
|
154
|
+
|
|
155
|
+
|
|
156
|
+
def test_paid_flag_survives_content_fallback() -> None:
|
|
157
|
+
html = _html("", data="window.cgiDataNew = {is_pay_subscribe: 1};")
|
|
158
|
+
html += '<div class="original_page"><p id="js_share_notice">转载正文</p></div>'
|
|
159
|
+
result = _parse_html(_URL, html)
|
|
160
|
+
|
|
161
|
+
assert result.article_is_paid is True
|
|
162
|
+
assert result.article_markdown == "转载正文"
|
|
163
|
+
|
|
164
|
+
|
|
165
|
+
def test_paid_flag_without_content_or_script_type() -> None:
|
|
166
|
+
result = _parse_html(_URL, "<script>var is_pay_subscribe = 1;</script>")
|
|
167
|
+
|
|
168
|
+
assert result.article_is_paid is True
|
|
169
|
+
assert not result.is_valid
|
|
170
|
+
|
|
171
|
+
|
|
172
|
+
def test_paid_preview_marker_in_repost() -> None:
|
|
173
|
+
result = _parse_html(
|
|
174
|
+
_URL, f'<div class="original_page"><p>转载预览</p>{_FILTER}</div>',
|
|
175
|
+
)
|
|
176
|
+
|
|
177
|
+
assert result.article_is_paid is True
|
|
178
|
+
|
|
179
|
+
|
|
180
|
+
def test_paid_preview_marker_in_script_repost_content() -> None:
|
|
181
|
+
html = (
|
|
182
|
+
'<div class="original_page"></div>'
|
|
183
|
+
'<script type="text/javascript">window.cgiDataNew = {'
|
|
184
|
+
"""content: '<p>预览</p><mp-pay-preview-filter></mp-pay-preview-filter>'"""
|
|
185
|
+
'};</script>'
|
|
186
|
+
)
|
|
187
|
+
result = _parse_html(_URL, html)
|
|
188
|
+
|
|
189
|
+
assert result.article_is_paid is True
|
|
190
|
+
assert result.article_markdown == "预览"
|
|
@@ -80,12 +80,14 @@ def _assert_result(result: ArticleResult, url: str) -> None:
|
|
|
80
80
|
else "公众号简介: (empty)"
|
|
81
81
|
)
|
|
82
82
|
print(f"文章ID: {result.article_id}")
|
|
83
|
+
print(f"链接类型: {result.article_url_type}")
|
|
83
84
|
print(f"群发消息ID: {result.article_msg_id}")
|
|
84
85
|
print(f"文章idx: {result.article_idx}")
|
|
85
86
|
print(f"文章签名: {result.article_sn}")
|
|
86
87
|
print(f"文章标题: {result.article_title}")
|
|
87
88
|
print(f"展示类型: {result.article_display_type}")
|
|
88
89
|
print(f"原始展示类型: {result.article_item_show_type}")
|
|
90
|
+
print(f"是否付费: {result.article_is_paid}")
|
|
89
91
|
print(
|
|
90
92
|
f"封面图: {result.article_cover_image[:80]}..."
|
|
91
93
|
if result.article_cover_image
|
|
@@ -133,12 +135,14 @@ def test_fetch_all(url: str, proxy: str | None) -> None:
|
|
|
133
135
|
print(f"公众号图片: {result.mp_image}")
|
|
134
136
|
print(f"公众号简介: {result.mp_description}")
|
|
135
137
|
print(f"文章ID: {result.article_id}")
|
|
138
|
+
print(f"链接类型: {result.article_url_type}")
|
|
136
139
|
print(f"群发消息ID: {result.article_msg_id}")
|
|
137
140
|
print(f"文章idx: {result.article_idx}")
|
|
138
141
|
print(f"文章签名: {result.article_sn}")
|
|
139
142
|
print(f"文章标题: {result.article_title}")
|
|
140
143
|
print(f"展示类型: {result.article_display_type}")
|
|
141
144
|
print(f"原始展示类型: {result.article_item_show_type}")
|
|
145
|
+
print(f"是否付费: {result.article_is_paid}")
|
|
142
146
|
print(f"封面图: {result.article_cover_image}")
|
|
143
147
|
print(f"文章摘要: {result.article_description}")
|
|
144
148
|
print(f"发布时间: {result.article_publish_time}")
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{wechat_article_parser-0.0.13 → wechat_article_parser-0.0.15}/tests/test_miniprogram_content.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|