first commit

.gitignore

@@ -0,0 +1,4 @@
+spider/chromedriver.exe
+spider/cache/
+__pycache__/
+.vscode/

README.md

@@ -0,0 +1,88 @@
+# pcr-wiki插件
+
+本插件需配合[Hoshino(v2)](https://github.com/Ice-Cirno/HoshinoBot)使用
+
+数据搬运于[蘭德索爾圖書館](https://pcredivewiki.tw/)，图片资源来源[干炸里脊资源站](https://redive.estertion.win/)
+
+## 功能
+
+- **[@bot简介ue] 角色简介**：查询角色简介
+- **[@bot技能ue] 角色技能**：查询角色技能
+- **[@bot技能ue] 角色专武**：查询角色专武
+- **[@bot羁绊ue] 角色羁绊**：查询角色羁绊
+- **启用wiki**：启用pcr-wiki
+- **禁用wiki**：禁用pcr-wiki
+
+## 部署
+
+1. 将本项目的`wiki`文件夹复制到`hoshino/modules/priconne`下
+
+2. 安装`requirements.txt`
+
+3. 将本项目的`skill`与`equipment`文件夹复制到`res/img/priconne`文件夹下面
+
+   > 实际上，只需要新建`skill`与`equipment`文件夹并把`skill`下的`icon_skill_ap01.png`、`icon_skill_ap02.png`、`icon_skill_attack.png`、`icon_skill_tack.png`四个图片复制过去就好，其他没有的图片使用时会自动下载
+
+4. 重启Hoshino
+
+   > 注意：**不要**在hoshino的配置文件添加模块。
+   >
+   > 注意：**不要**把`spider`文件夹及该文件夹下的文件任何文件放到`hoshino`下
+
+至此，你可以开始使用插件了。
+
+插件的数据源自文件夹下的`data.db`，wiki有更新时需要对应更新（我尝试过实时拉取数据，速度太慢，所以改用数据库存储），`data.db`会不定时更新，如果你想要自己手动更新，请看下一小节
+
+## 更新数据
+
+> 强烈建议在windows机器上更新数据，更为快速方便。`spider`文件夹仅作更新数据使用，**不要**把这个文件夹混入`hoshino`的任何目录，它是独立的
+
+#### windows
+
+1. 打开`spider`文件夹，安装`requirements.txt`
+
+2. 将你需要更新的`data.db`准备好
+
+3. 安装chrome浏览器，并查看chrome版本
+
+4. http://npm.taobao.org/mirrors/chromedriver/下载最为接近你的chrome版本的驱动
+
+   > **不要**双击运行解压得到的exe文件，看下一步！
+
+5. 打开`run.py`按照注释修改对应处（第24、27、35行），打开`data.py`按照注释修改第3行
+
+6. 将你最新的`_pcr_data.py`复制到`spider`文件夹下替换（保证`spider/_pcr_data.py`里有你需要更新的id信息）
+
+7. 运行`run.py`
+
+8. 若无报错，则更新成功，得到最新的`data.db`，替换掉你`hoshino/modules/priconne/wiki`下的同名件
+
+#### Linux
+
+1. 将`spider`文件夹复制到服务器，打开该文件夹，安装`requirments.txt`
+
+2. 将你需要更新的`data.db`准备好
+
+3. 安装chrome浏览器，并查看chrome版本
+
+   依次运行以下命令（第二行大概率会报错，不用管，但必须运行）
+
+   ```
+   wget https://dl.google.com/linux/direct/google-chrome-stable_current_amd64.deb
+   sudo dpkg -i google-chrome-stable_current_amd64.deb
+   sudo apt-get install -f
+   sudo dpkg -i google-chrome-stable_current_amd64.deb
+   sudo apt-get install xvfb
+   google-chrome --version
+   ```
+
+4. http://npm.taobao.org/mirrors/chromedriver/下载最为接近你的chrome版本的驱动
+
+5. 打开`run.py`按照注释修改对应处（第24、27、35行），打开`data.py`按照注释修改第3行
+
+6. 将你最新的`_pcr_data.py`复制到`spider`文件夹下替换（保证`spider/_pcr_data.py`里有你需要更新的id信息）
+
+7. 运行`run.py`
+
+8. 若无报错，则更新成功，得到最新的`data.db`，替换掉你`hoshino/modules/priconne/wiki`下的同名件
+

spider/Fetch.py

@@ -0,0 +1,184 @@
+import re, time, os
+from selenium.common.exceptions import NoSuchElementException
+def get_chara_skill(driver, research=0):
+    try:
+        skill_boxes = driver.find_elements_by_class_name('skill-box')
+        chara_skells = []
+        for s_b in skill_boxes:
+            skill = {}
+            skill['skill_name'] = s_b.find_element_by_class_name('skill-name').text
+            skill['skill_type'] = s_b.find_element_by_class_name('skill-type').text
+
+            skill['description'] = s_b.find_element_by_class_name('skill-description').text
+            skill['description'] = re.sub(r'(EX)?技能[1-3]?\+?|必殺技|\n','', skill['description'])
+            skill['description'] = skill['description'].replace(skill['skill_name'], '')
+
+            img_src = s_b.find_element_by_class_name('img-fluid').get_attribute('src')
+            num_match = re.search(r'[0-9]+', os.path.basename(img_src))
+            if num_match:
+                skill['skill_num'] = num_match.group(0)
+            else:
+                skill['skill_num'] = -1
+
+            skill['skill_effect'] = []
+            skill_effects = s_b.find_elements_by_css_selector('.skill-effect > .mb-2')
+            for s_e in skill_effects:
+                s_e_divs = s_e.find_elements_by_tag_name('div')
+                e_txt = ''
+                for div in s_e_divs:
+                    if div.text:
+                        e_txt += div.text
+                if e_txt:
+                    e_txt = e_txt.replace('\n', '')
+                    skill['skill_effect'].append(e_txt)
+            chara_skells.append(skill)
+        return chara_skells
+    except NoSuchElementException as err :
+        research += 1
+        print(f'get_chara_skill_error ({research}) : {err}')
+        time.sleep(1)
+        if research < 4:
+            get_chara_skill(driver, research )
+        else:
+            raise NoSuchElementException(err) 
+
+def get_chara_info(driver, research=0):
+    info = {}
+    keys = {
+        '名字': 'name',
+        '公會': 'guild',
+        '生日': 'birthday',
+        '年齡': 'age',
+        '身高': 'height',
+        '體重': 'weight',
+        '血型': 'blood_type',
+        '種族': 'race',
+        '喜好': 'hobby',
+        '聲優': 'cv'
+    }
+    try:
+        # 获得角色信息
+        table = driver.find_element_by_class_name('chara-table')
+        tr_list = table.find_elements_by_tag_name('tr')
+        for tr in tr_list:
+            item = tr.find_element_by_tag_name('th').text
+            info[keys.get(item, item)] = tr.find_element_by_tag_name('td').text
+
+        # 获取介绍
+        intro = driver.find_elements_by_css_selector('.row>.col-lg-7>.d-block')
+        if intro:
+            info['introduce']= intro[0].text
+            info['introduce'] = info['introduce'].replace('簡介\n', '')
+        return info
+    except NoSuchElementException as err:
+        research += 1
+        print(f'get_chara_info_error ({research}) : {err}')
+        time.sleep(1)
+        if research < 3:
+            get_chara_info(driver, research)
+        else:
+            raise NoSuchElementException(err)
+
+def get_chara_kizuna(driver):
+    kizuna_tables = driver.find_elements_by_css_selector('.chara-table.mb-3')
+    count = len(kizuna_tables)
+    if count != 1: count = count / 3
+    kizuna = {}
+    for table in kizuna_tables:
+        # 检查是否爬取到所有羁绊
+        if len(kizuna) == count:
+            break
+
+        name = table.find_element_by_xpath('.//th[@colspan=2]').get_attribute('innerText')
+        # 检查羁绊是否以及爬取过
+        if name in kizuna or name == '':
+            continue
+        else:
+            # 创建角色羁绊
+            kizuna[name] = {}
+
+        for tr in table.find_elements_by_tag_name('tr')[1:]:
+
+            epi_key = tr.find_element_by_tag_name('th').get_attribute('innerText')
+            effect = tr.find_element_by_tag_name('td').get_attribute('innerHTML')
+
+            # html标签替换为分割符
+            effect = re.sub(r'<div[^>]*>|</div>|<!-+>', '~', effect)
+            # 利用贪婪匹配去除多余分割符
+            effect = re.sub(r'~+', '~', effect)
+            effect = effect.strip('~')
+            kizuna[name][epi_key] = effect.split('~')
+    return kizuna
+
+def get_act_pattern(driver):
+    act_pattern = {}
+    for h4 in driver.find_elements_by_css_selector('h4.item-sub-title'):
+        if h4.text == '起手':
+            parent_node = h4.find_element_by_xpath('..')
+            start_pattern = []
+            for img in parent_node.find_elements_by_tag_name('img'):
+                img_src = img.get_attribute('src')
+                pt = os.path.splitext(os.path.basename(img_src))[0][-4:]
+                start_pattern.append(pt)
+            act_pattern['start'] = start_pattern
+
+        if h4.text == '循環':
+            parent_node = h4.find_element_by_xpath('..')
+            loop_pattern = []
+            for img in parent_node.find_elements_by_tag_name('img'):
+                img_src = img.get_attribute('src')
+                act = os.path.splitext(os.path.basename(img_src))[0][-4:]
+                if act == 'tack':
+                    act = 'attack'
+                loop_pattern.append(act)
+            act_pattern['loop'] = loop_pattern   
+
+    return act_pattern
+
+def get_unique_weapon(driver):
+    for h3 in driver.find_elements_by_css_selector('h3.item-title.mb-0'):
+        weapon = {}
+        if '專用裝備' in h3.text:
+            box = h3.find_element_by_xpath('..').find_element_by_class_name('prod-info-box')
+            imgsrc = box.find_element_by_class_name('img-fluid').get_attribute('src')
+            num_match = re.search(r'[0-9]+', os.path.basename(imgsrc))
+            if num_match:
+                weapon['weapon_num'] = num_match.group(0)
+            else:
+                weapon['weapon_num'] = -1
+
+            weapon['name'] = box.find_element_by_id('0').text
+            weapon['description'] = box.find_element_by_css_selector('.col-lg-8.col-12>p').text
+            weapon['props'] = []
+            for pop_box in box.find_elements_by_class_name('prod-data'):
+                prop = {}
+                prop['property'] = pop_box.find_element_by_class_name('title').text
+                value = pop_box.find_element_by_class_name('prod-value').text
+                value = value.replace(' ', '')
+                value = value.replace(')', '')
+                value_list = value.split('(')
+                prop['base_value'] = value_list[0]
+                prop['max_value'] = value_list[1]
+                weapon['props'].append(prop)
+            return weapon if weapon else False 
+
+def fetch(driver):
+    chara_data = {}
+    chara_data['info'] = get_chara_info(driver)
+    chara_data['kizuna'] = get_chara_kizuna(driver)
+    chara_data['skill'] = get_chara_skill(driver)
+
+    uniquei_weapon = get_unique_weapon(driver)
+    if uniquei_weapon:
+        # skills = chara_data['skill']
+        # for index in range(len(skills)):
+        #     if '專武強化技能' in skills[index]['skill_type']:
+        #         uniquei_weapon['strengthen_skill'] = skills[index]
+        #         # 删除技能中的专武强化技
+        #         # skills.pop(index)
+        #         break
+
+        chara_data['uniquei'] = uniquei_weapon
+
+    chara_data['pattern'] = get_act_pattern(driver)
+    return chara_data