p15679402
/
bigdata
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
bigdata/chapter1/.idea/workspace.xml

676 lines
33 KiB
XML
Raw Permalink Normal View History

抓取评论
2019-01-13 11:34:47 +08:00
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="ChangeListManager">
web
2019-01-17 20:15:22 +08:00
<list default="true" id="a5fbb387-9969-4874-8a7f-c9dd40d5225d" name="Default Changelist" comment="" />
抓取评论
2019-01-13 11:34:47 +08:00
<option name="EXCLUDED_CONVERTED_TO_IGNORED" value="true" />
<option name="SHOW_DIALOG" value="false" />
<option name="HIGHLIGHT_CONFLICTS" value="true" />
<option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
<option name="LAST_RESOLUTION" value="IGNORE" />
</component>
<component name="CoverageDataManager">
<SUITE FILE_PATH="coverage/chapter1$category.coverage" NAME="category Coverage Results" MODIFIED="1546938251328" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="true" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/crawler/taobao/spiders" />
<SUITE FILE_PATH="coverage/chapter1$merge.coverage" NAME="merge Coverage Results" MODIFIED="1547014157363" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="true" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/crawler/taobao" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<SUITE FILE_PATH="coverage/chapter1$demo.coverage" NAME="demo Coverage Results" MODIFIED="1547519456164" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="true" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/crawler" />
抓取评论
2019-01-13 11:34:47 +08:00
<SUITE FILE_PATH="coverage/chapter1$comment.coverage" NAME="comment Coverage Results" MODIFIED="1547166377597" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="true" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/crawler/taobao/spiders" />
</component>
<component name="FileEditorManager">
<leaf SIDE_TABS_SIZE_LIMIT_KEY="300">
<file pinned="false" current-in-tab="false">
<entry file="file://$PROJECT_DIR$/crawler/taobao/items.py">
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="238">
<caret line="64" column="14" selection-start-line="64" selection-start-column="4" selection-end-line="64" selection-end-column="14" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
</file>
<file pinned="false" current-in-tab="false">
导入hbase
2019-01-14 22:34:53 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/pipelines.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="-1250">
<caret line="15" lean-forward="true" selection-start-line="15" selection-end-line="15" />
导入hbase
2019-01-14 22:34:53 +08:00
<folding>
<element signature="e#193#204#0" expanded="true" />
</folding>
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
</file>
加入京东抓取
2019-01-14 16:28:26 +08:00
<file pinned="false" current-in-tab="false">
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/utils.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="60">
<caret line="4" column="12" selection-start-line="4" selection-start-column="4" selection-end-line="4" selection-end-column="12" />
导入hbase
2019-01-14 22:34:53 +08:00
</state>
</provider>
</entry>
</file>
<file pinned="false" current-in-tab="false">
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/settings.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="101">
<caret line="110" column="28" lean-forward="true" selection-start-line="110" selection-start-column="28" selection-end-line="110" selection-end-column="28" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
</file>
开始第二章
2019-01-15 19:06:36 +08:00
<file pinned="false" current-in-tab="false">
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/client.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="235">
开始第二章
2019-01-15 19:06:36 +08:00
<caret line="31" column="31" lean-forward="true" selection-start-line="31" selection-start-column="31" selection-end-line="31" selection-end-column="31" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
</file>
开始第二章
2019-01-15 19:06:36 +08:00
<file pinned="false" current-in-tab="true">
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/jd.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
web
2019-01-17 20:15:22 +08:00
<state relative-caret-position="242">
<caret line="140" column="22" selection-start-line="140" selection-start-column="22" selection-end-line="140" selection-end-column="22" />
抓取评论
2019-01-13 11:34:47 +08:00
<folding>
抓取京东评论
2019-01-15 16:07:01 +08:00
<element signature="e#15#26#0" expanded="true" />
抓取评论
2019-01-13 11:34:47 +08:00
</folding>
</state>
</provider>
</entry>
</file>
<file pinned="false" current-in-tab="false">
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/http/request/__init__.py">
加入京东抓取
2019-01-14 16:28:26 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="180">
<caret line="19" column="68" selection-start-line="19" selection-start-column="60" selection-end-line="19" selection-end-column="68" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
</file>
</leaf>
</component>
<component name="FileTemplateManagerImpl">
<option name="RECENT_TEMPLATES">
<list>
<option value="Python Script" />
</list>
</option>
</component>
<component name="FindInProjectRecents">
<findStrings>
<find>日本可爱毛球熊猫束发带</find>
<find>美式风</find>
<find>雪</find>
<find>皮床</find>
加入京东抓取
2019-01-14 16:28:26 +08:00
<find>SPLASH</find>
<find>DOWNLOADER_MIDDLEWARES</find>
<find>DUPEFILTER_CLASS</find>
<find>contains</find>
<find>scrapy.selector.unified.Selector</find>
抓取京东评论
2019-01-15 16:07:01 +08:00
<find>AUTOTHROTTLE_ENABLED</find>
<find>WEBSERVICE_PORT</find>
<find>WEBSERVICE_ENABLED</find>
<find>TELNETCONSOLE_ENABLED</find>
<find>EXTENSIONS_BASE</find>
<find>TELNET</find>
开始第二章
2019-01-15 19:06:36 +08:00
<find>HBASE_PORT</find>
抓取评论
2019-01-13 11:34:47 +08:00
</findStrings>
</component>
<component name="Git.Settings">
<option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$/.." />
</component>
<component name="IdeDocumentHistory">
<option name="CHANGED_PATHS">
<list>
<option value="$PROJECT_DIR$/crawler/taobao/middlewares.py" />
<option value="$PROJECT_DIR$/crawler/taobao/run.py" />
<option value="$PROJECT_DIR$/crawler/taobao/spiders/dangdang.py" />
<option value="$PROJECT_DIR$/crawler/taobao/spiders/run.py" />
<option value="$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/spiders/__init__.py" />
<option value="$PROJECT_DIR$/crawler/datas/kind.json" />
<option value="$PROJECT_DIR$/crawler/taobao/merge.py" />
<option value="$PROJECT_DIR$/crawler/taobao/spiders/comment.py" />
加入京东抓取
2019-01-14 16:28:26 +08:00
<option value="$PROJECT_DIR$/crawler/taobao/spiders/category.py" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<option value="$PROJECT_DIR$/crawler/demo.py" />
<option value="$PROJECT_DIR$/crawler/taobao/utils.py" />
加入京东抓取
2019-01-14 16:28:26 +08:00
<option value="$PROJECT_DIR$/crawler/taobao/items.py" />
<option value="$PROJECT_DIR$/crawler/taobao/pipelines.py" />
<option value="$PROJECT_DIR$/crawler/taobao/spiders/jd.py" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<option value="$PROJECT_DIR$/crawler/taobao/settings.py" />
<option value="$PROJECT_DIR$/crawler/client.py" />
抓取评论
2019-01-13 11:34:47 +08:00
</list>
</option>
</component>
<component name="ProjectFrameBounds">
<option name="y" value="23" />
<option name="width" value="1280" />
<option name="height" value="777" />
</component>
<component name="ProjectLevelVcsManager" settingsEditedManually="true" />
<component name="ProjectView">
<navigator proportions="" version="1">
<foldersAlwaysOnTop value="true" />
</navigator>
<panes>
抓取京东评论
2019-01-15 16:07:01 +08:00
<pane id="Scope" />
抓取评论
2019-01-13 11:34:47 +08:00
<pane id="ProjectPane">
<subPane>
<expand>
<path>
<item name="chapter1" type="b2602c69:ProjectViewProjectNode" />
<item name="chapter1" type="462c0819:PsiDirectoryNode" />
</path>
</expand>
<select />
</subPane>
</pane>
</panes>
</component>
<component name="PropertiesComponent">
<property name="WebServerToolWindowFactoryState" value="false" />
<property name="last_opened_file_path" value="$PROJECT_DIR$" />
<property name="nodejs_interpreter_path.stuck_in_default_project" value="undefined stuck path" />
<property name="nodejs_npm_path_reset_for_default_project" value="true" />
加入京东抓取
2019-01-14 16:28:26 +08:00
<property name="settings.editor.selected.configurable" value="com.jetbrains.python.configuration.PyActiveSdkModuleConfigurable" />
抓取评论
2019-01-13 11:34:47 +08:00
</component>
<component name="RecentsManager">
<key name="MoveFile.RECENT_KEYS">
<recent name="$PROJECT_DIR$/crawler/taobao/spiders" />
</key>
</component>
<component name="RunDashboard">
<option name="ruleStates">
<list>
<RuleState>
<option name="name" value="ConfigurationTypeDashboardGroupingRule" />
</RuleState>
<RuleState>
<option name="name" value="StatusDashboardGroupingRule" />
</RuleState>
</list>
</option>
</component>
抓取京东评论
2019-01-15 16:07:01 +08:00
<component name="RunManager" selected="Python.demo">
抓取评论
2019-01-13 11:34:47 +08:00
<configuration name="category" type="PythonConfigurationType" factoryName="Python" temporary="true">
<module name="chapter1" />
<option name="INTERPRETER_OPTIONS" value="" />
<option name="PARENT_ENVS" value="true" />
<envs>
<env name="PYTHONUNBUFFERED" value="1" />
</envs>
<option name="SDK_HOME" value="" />
<option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/crawler/taobao/spiders" />
<option name="IS_MODULE_SDK" value="true" />
<option name="ADD_CONTENT_ROOTS" value="true" />
<option name="ADD_SOURCE_ROOTS" value="true" />
<EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
<option name="SCRIPT_NAME" value="$PROJECT_DIR$/crawler/taobao/spiders/category.py" />
<option name="PARAMETERS" value="" />
<option name="SHOW_COMMAND_LINE" value="false" />
<option name="EMULATE_TERMINAL" value="false" />
<option name="MODULE_MODE" value="false" />
<option name="REDIRECT_INPUT" value="false" />
<option name="INPUT_FILE" value="" />
<method v="2" />
</configuration>
<configuration name="comment" type="PythonConfigurationType" factoryName="Python" temporary="true">
<module name="chapter1" />
<option name="INTERPRETER_OPTIONS" value="" />
<option name="PARENT_ENVS" value="true" />
<envs>
<env name="PYTHONUNBUFFERED" value="1" />
</envs>
<option name="SDK_HOME" value="" />
<option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/crawler/taobao/spiders" />
<option name="IS_MODULE_SDK" value="true" />
<option name="ADD_CONTENT_ROOTS" value="true" />
<option name="ADD_SOURCE_ROOTS" value="true" />
<EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
<option name="SCRIPT_NAME" value="$PROJECT_DIR$/crawler/taobao/spiders/comment.py" />
<option name="PARAMETERS" value="" />
<option name="SHOW_COMMAND_LINE" value="false" />
<option name="EMULATE_TERMINAL" value="false" />
<option name="MODULE_MODE" value="false" />
<option name="REDIRECT_INPUT" value="false" />
<option name="INPUT_FILE" value="" />
<method v="2" />
</configuration>
抓取京东评论
2019-01-15 16:07:01 +08:00
<configuration name="demo" type="PythonConfigurationType" factoryName="Python" temporary="true">
<module name="chapter1" />
<option name="INTERPRETER_OPTIONS" value="" />
<option name="PARENT_ENVS" value="true" />
<envs>
<env name="PYTHONUNBUFFERED" value="1" />
</envs>
<option name="SDK_HOME" value="" />
<option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/crawler" />
<option name="IS_MODULE_SDK" value="true" />
<option name="ADD_CONTENT_ROOTS" value="true" />
<option name="ADD_SOURCE_ROOTS" value="true" />
<EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
<option name="SCRIPT_NAME" value="$PROJECT_DIR$/crawler/demo.py" />
<option name="PARAMETERS" value="" />
<option name="SHOW_COMMAND_LINE" value="false" />
<option name="EMULATE_TERMINAL" value="false" />
<option name="MODULE_MODE" value="false" />
<option name="REDIRECT_INPUT" value="false" />
<option name="INPUT_FILE" value="" />
<method v="2" />
</configuration>
抓取评论
2019-01-13 11:34:47 +08:00
<configuration name="merge" type="PythonConfigurationType" factoryName="Python" temporary="true">
<module name="chapter1" />
<option name="INTERPRETER_OPTIONS" value="" />
<option name="PARENT_ENVS" value="true" />
<envs>
<env name="PYTHONUNBUFFERED" value="1" />
</envs>
<option name="SDK_HOME" value="" />
<option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/crawler/taobao" />
<option name="IS_MODULE_SDK" value="true" />
<option name="ADD_CONTENT_ROOTS" value="true" />
<option name="ADD_SOURCE_ROOTS" value="true" />
<EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
<option name="SCRIPT_NAME" value="$PROJECT_DIR$/crawler/taobao/merge.py" />
<option name="PARAMETERS" value="" />
<option name="SHOW_COMMAND_LINE" value="false" />
<option name="EMULATE_TERMINAL" value="false" />
<option name="MODULE_MODE" value="false" />
<option name="REDIRECT_INPUT" value="false" />
<option name="INPUT_FILE" value="" />
<method v="2" />
</configuration>
<recent_temporary>
<list>
抓取京东评论
2019-01-15 16:07:01 +08:00
<item itemvalue="Python.demo" />
抓取评论
2019-01-13 11:34:47 +08:00
<item itemvalue="Python.comment" />
<item itemvalue="Python.merge" />
<item itemvalue="Python.category" />
</list>
</recent_temporary>
</component>
<component name="TaskManager">
<task active="true" id="Default" summary="Default task">
<changelist id="a5fbb387-9969-4874-8a7f-c9dd40d5225d" name="Default Changelist" comment="" />
<created>1546928387088</created>
<option name="number" value="Default" />
<option name="presentableId" value="Default" />
<updated>1546928387088</updated>
</task>
<task id="LOCAL-00001" summary="数据抓取">
<created>1546936657398</created>
<option name="number" value="00001" />
<option name="presentableId" value="LOCAL-00001" />
<option name="project" value="LOCAL" />
<updated>1546936657399</updated>
</task>
<task id="LOCAL-00002" summary="增加id字段">
<created>1546938314632</created>
<option name="number" value="00002" />
<option name="presentableId" value="LOCAL-00002" />
<option name="project" value="LOCAL" />
<updated>1546938314632</updated>
</task>
.
2019-01-14 16:40:25 +08:00
<task id="LOCAL-00003" summary="加入京东抓取">
<created>1547454506424</created>
<option name="number" value="00003" />
<option name="presentableId" value="LOCAL-00003" />
<option name="project" value="LOCAL" />
<updated>1547454506424</updated>
</task>
抓取京东评论
2019-01-15 16:07:01 +08:00
<task id="LOCAL-00004" summary="导入hbase">
<created>1547476493838</created>
<option name="number" value="00004" />
<option name="presentableId" value="LOCAL-00004" />
<option name="project" value="LOCAL" />
<updated>1547476493838</updated>
</task>
开始第二章
2019-01-15 19:06:36 +08:00
<task id="LOCAL-00005" summary="开始第二章">
<created>1547550283596</created>
<option name="number" value="00005" />
<option name="presentableId" value="LOCAL-00005" />
<option name="project" value="LOCAL" />
<updated>1547550283596</updated>
</task>
web
2019-01-17 20:15:22 +08:00
<task id="LOCAL-00006" summary="开始第二章">
<created>1547550396383</created>
<option name="number" value="00006" />
<option name="presentableId" value="LOCAL-00006" />
<option name="project" value="LOCAL" />
<updated>1547550396383</updated>
</task>
<option name="localTasksCounter" value="7" />
抓取评论
2019-01-13 11:34:47 +08:00
<servers />
</component>
<component name="TodoView">
<todo-panel id="selected-file">
<is-autoscroll-to-source value="true" />
</todo-panel>
<todo-panel id="all">
<are-packages-shown value="true" />
<is-autoscroll-to-source value="true" />
</todo-panel>
</component>
<component name="ToolWindowManager">
<frame x="0" y="23" width="1280" height="777" extended-state="0" />
<layout>
web
2019-01-17 20:15:22 +08:00
<window_info active="true" content_ui="combo" id="Project" order="0" visible="true" weight="0.26171243" />
抓取评论
2019-01-13 11:34:47 +08:00
<window_info id="Structure" order="1" side_tool="true" weight="0.25" />
<window_info id="Favorites" order="2" side_tool="true" />
<window_info anchor="bottom" id="Message" order="0" />
<window_info anchor="bottom" id="Find" order="1" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<window_info anchor="bottom" id="Run" order="2" weight="0.2729927" />
抓取评论
2019-01-13 11:34:47 +08:00
<window_info anchor="bottom" id="Debug" order="3" weight="0.4" />
<window_info anchor="bottom" id="Cvs" order="4" weight="0.25" />
<window_info anchor="bottom" id="Inspection" order="5" weight="0.4" />
<window_info anchor="bottom" id="TODO" order="6" weight="0.329927" />
<window_info anchor="bottom" id="Version Control" order="7" />
web
2019-01-17 20:15:22 +08:00
<window_info anchor="bottom" id="Terminal" order="8" visible="true" weight="0.24525547" />
抓取评论
2019-01-13 11:34:47 +08:00
<window_info anchor="bottom" id="Event Log" order="9" side_tool="true" />
<window_info anchor="bottom" id="Python Console" order="10" weight="0.329927" />
<window_info anchor="right" id="Commander" internal_type="SLIDING" order="0" type="SLIDING" weight="0.4" />
<window_info anchor="right" id="Ant Build" order="1" weight="0.25" />
<window_info anchor="right" content_ui="combo" id="Hierarchy" order="2" weight="0.25" />
<window_info anchor="right" id="SciView" order="3" />
</layout>
</component>
<component name="VcsManagerConfiguration">
<MESSAGE value="数据抓取" />
<MESSAGE value="增加id字段" />
.
2019-01-14 16:40:25 +08:00
<MESSAGE value="加入京东抓取" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<MESSAGE value="导入hbase" />
开始第二章
2019-01-15 19:06:36 +08:00
<MESSAGE value="开始第二章" />
<option name="LAST_COMMIT_MESSAGE" value="开始第二章" />
抓取评论
2019-01-13 11:34:47 +08:00
</component>
<component name="editorHistoryManager">
<entry file="file://$PROJECT_DIR$/crawler/scrapy.cfg">
<provider selected="true" editor-type-id="text-editor" />
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/users.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="255">
<caret line="18" column="52" selection-start-line="18" selection-start-column="52" selection-end-line="18" selection-end-column="52" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/utils/project.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="140">
<caret line="59" column="4" selection-start-line="59" selection-start-column="4" selection-end-line="59" selection-end-column="4" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/run.py" />
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/spiders/__init__.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="238">
<caret line="32" column="32" selection-start-line="32" selection-start-column="32" selection-end-line="32" selection-end-column="32" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/datas/大数据.json" />
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/cmdline.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="1365">
<caret line="105" column="36" selection-start-line="105" selection-start-column="28" selection-end-line="105" selection-end-column="36" />
</state>
</provider>
</entry>
<entry file="file:///usr/local/Cellar/python/3.7.0/Frameworks/Python.framework/Versions/3.7/lib/python3.7/json/__init__.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="146">
<caret line="356" column="29" selection-start-line="356" selection-start-column="29" selection-end-line="356" selection-end-column="29" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/datas/T恤.json" />
<entry file="file://$PROJECT_DIR$/crawler/datas/冬上新.json" />
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/crawler.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="-93">
<caret line="147" column="8" selection-start-line="147" selection-start-column="8" selection-end-line="147" selection-end-column="8" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/cssselect/xpath.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="118">
<caret line="247" column="8" selection-start-line="247" selection-start-column="8" selection-end-line="247" selection-end-column="8" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/http/response/__init__.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="110">
<caret line="18" column="26" selection-start-line="18" selection-start-column="23" selection-end-line="18" selection-end-column="26" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/../../python/crawler/env/lib/python3.7/site-packages/scrapy/http/response/__init__.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="130">
<caret line="54" column="35" lean-forward="true" selection-start-line="54" selection-start-column="35" selection-end-line="54" selection-end-column="35" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/datas/内衣套装.json" />
<entry file="file://$PROJECT_DIR$/../../python/crawler/env/lib/python3.7/site-packages/scrapy/item.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="361">
<caret line="72" selection-start-line="72" selection-end-line="72" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/hah.json" />
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/douban.json" />
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/datas/comments/23761145.json" />
抓取评论
2019-01-13 11:34:47 +08:00
<entry file="file://$PROJECT_DIR$/crawler/datas/2小时送货服务.json" />
<entry file="file://$PROJECT_DIR$/crawler/datas/3万以下.json" />
<entry file="file://$PROJECT_DIR$/crawler/datas/kind.json" />
<entry file="file://$PROJECT_DIR$/crawler/datas/products/data.json" />
<entry file="file://$USER_HOME$/Library/Caches/PyCharm2018.3/python_stubs/-1583084946/builtins.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="36915">
<caret line="2461" column="42" selection-start-line="2461" selection-start-column="34" selection-end-line="2461" selection-end-column="42" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/jupyter.py">
<provider selected="true" editor-type-id="text-editor" />
</entry>
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file:///usr/local/Cellar/python/3.7.0/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
加入京东抓取
2019-01-14 16:28:26 +08:00
<state relative-caret-position="139">
<caret line="965" column="8" selection-start-line="965" selection-start-column="8" selection-end-line="965" selection-end-column="8" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/lxml/__init__.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
加入京东抓取
2019-01-14 16:28:26 +08:00
<state relative-caret-position="300">
<caret line="20" lean-forward="true" selection-start-line="20" selection-end-line="20" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/httpbin.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
加入京东抓取
2019-01-14 16:28:26 +08:00
<state relative-caret-position="45">
<caret line="3" column="13" selection-start-line="3" selection-start-column="5" selection-end-line="3" selection-end-column="13" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/dangdang.py">
<provider selected="true" editor-type-id="text-editor">
加入京东抓取
2019-01-14 16:28:26 +08:00
<state relative-caret-position="-318">
抓取评论
2019-01-13 11:34:47 +08:00
<caret line="21" column="22" selection-start-line="21" selection-start-column="22" selection-end-line="21" selection-end-column="22" />
<folding>
<element signature="e#119#129#0" expanded="true" />
</folding>
</state>
</provider>
</entry>
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/category.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="251">
<caret line="47" column="59" selection-start-line="47" selection-start-column="54" selection-end-line="47" selection-end-column="59" />
<folding>
<element signature="e#24#35#0" expanded="true" />
</folding>
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/parsel/selector.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="245">
导入hbase
2019-01-14 22:34:53 +08:00
<caret line="238" column="39" selection-start-line="238" selection-start-column="39" selection-end-line="238" selection-end-column="39" />
加入京东抓取
2019-01-14 16:28:26 +08:00
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/selector/unified.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="492">
<caret line="84" column="24" selection-start-line="84" selection-start-column="8" selection-end-line="84" selection-end-column="24" />
</state>
</provider>
</entry>
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/conf.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="105">
<caret line="7" column="4" selection-start-line="7" selection-start-column="4" selection-end-line="7" selection-end-column="4" />
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/happybase/connection.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="1455">
<caret line="107" column="8" selection-start-line="107" selection-start-column="8" selection-end-line="107" selection-end-column="8" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
导入hbase
2019-01-14 22:34:53 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/comment.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="420">
导入hbase
2019-01-14 22:34:53 +08:00
<caret line="36" column="53" selection-start-line="36" selection-start-column="53" selection-end-line="36" selection-end-column="53" />
<folding>
<element signature="e#24#35#0" expanded="true" />
</folding>
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/middlewares.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="1920">
<caret line="137" column="32" selection-start-line="137" selection-start-column="32" selection-end-line="137" selection-end-column="32" />
<folding>
<element signature="e#168#178#0" expanded="true" />
</folding>
</state>
</provider>
</entry>
加入京东抓取
2019-01-14 16:28:26 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/item.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="162">
<caret line="58" selection-start-line="58" selection-end-line="58" />
加入京东抓取
2019-01-14 16:28:26 +08:00
</state>
</provider>
</entry>
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/__init__.py">
<provider selected="true" editor-type-id="text-editor" />
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/merge.py">
加入京东抓取
2019-01-14 16:28:26 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="345">
<caret line="23" column="30" selection-start-line="23" selection-start-column="30" selection-end-line="23" selection-end-column="30" />
加入京东抓取
2019-01-14 16:28:26 +08:00
</state>
</provider>
</entry>
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy_splash/request.py">
加入京东抓取
2019-01-14 16:28:26 +08:00
<provider selected="true" editor-type-id="text-editor">
抓取京东评论
2019-01-15 16:07:01 +08:00
<state relative-caret-position="-632">
<caret line="20" column="8" selection-start-line="20" selection-start-column="8" selection-end-line="20" selection-end-column="8" />
抓取评论
2019-01-13 11:34:47 +08:00
</state>
</provider>
</entry>
抓取京东评论
2019-01-15 16:07:01 +08:00
<entry file="file://$PROJECT_DIR$/crawler/demo.py">
<provider selected="true" editor-type-id="text-editor">
<state relative-caret-position="311">
<caret line="26" column="36" lean-forward="true" selection-start-line="26" selection-start-column="36" selection-end-line="26" selection-end-column="36" />
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file:///usr/local/Cellar/python/3.7.0/Frameworks/Python.framework/Versions/3.7/lib/python3.7/telnetlib.py">
抓取评论
2019-01-13 11:34:47 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="291">
<caret line="289" column="33" lean-forward="true" selection-start-line="289" selection-start-column="33" selection-end-line="289" selection-end-column="33" />
抓取京东评论
2019-01-15 16:07:01 +08:00
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file://$PROJECT_DIR$/env/lib/python3.7/site-packages/scrapy/http/request/__init__.py">
抓取京东评论
2019-01-15 16:07:01 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="180">
<caret line="19" column="68" selection-start-line="19" selection-start-column="60" selection-end-line="19" selection-end-column="68" />
抓取京东评论
2019-01-15 16:07:01 +08:00
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/items.py">
抓取京东评论
2019-01-15 16:07:01 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="238">
<caret line="64" column="14" selection-start-line="64" selection-start-column="4" selection-end-line="64" selection-end-column="14" />
导入hbase
2019-01-14 22:34:53 +08:00
</state>
</provider>
</entry>
<entry file="file://$PROJECT_DIR$/crawler/taobao/pipelines.py">
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="-1250">
<caret line="15" lean-forward="true" selection-start-line="15" selection-end-line="15" />
抓取评论
2019-01-13 11:34:47 +08:00
<folding>
导入hbase
2019-01-14 22:34:53 +08:00
<element signature="e#193#204#0" expanded="true" />
抓取评论
2019-01-13 11:34:47 +08:00
</folding>
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/utils.py">
抓取京东评论
2019-01-15 16:07:01 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="60">
<caret line="4" column="12" selection-start-line="4" selection-start-column="4" selection-end-line="4" selection-end-column="12" />
抓取京东评论
2019-01-15 16:07:01 +08:00
</state>
</provider>
</entry>
导入hbase
2019-01-14 22:34:53 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/settings.py">
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="101">
<caret line="110" column="28" lean-forward="true" selection-start-line="110" selection-start-column="28" selection-end-line="110" selection-end-column="28" />
导入hbase
2019-01-14 22:34:53 +08:00
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file://$PROJECT_DIR$/crawler/client.py">
加入京东抓取
2019-01-14 16:28:26 +08:00
<provider selected="true" editor-type-id="text-editor">
开始第二章
2019-01-15 19:06:36 +08:00
<state relative-caret-position="235">
<caret line="31" column="31" lean-forward="true" selection-start-line="31" selection-start-column="31" selection-end-line="31" selection-end-column="31" />
抓取京东评论
2019-01-15 16:07:01 +08:00
</state>
</provider>
</entry>
开始第二章
2019-01-15 19:06:36 +08:00
<entry file="file://$PROJECT_DIR$/crawler/taobao/spiders/jd.py">
抓取京东评论
2019-01-15 16:07:01 +08:00
<provider selected="true" editor-type-id="text-editor">
web
2019-01-17 20:15:22 +08:00
<state relative-caret-position="242">
<caret line="140" column="22" selection-start-line="140" selection-start-column="22" selection-end-line="140" selection-end-column="22" />
开始第二章
2019-01-15 19:06:36 +08:00
<folding>
<element signature="e#15#26#0" expanded="true" />
</folding>
加入京东抓取
2019-01-14 16:28:26 +08:00
</state>
</provider>
</entry>
抓取评论
2019-01-13 11:34:47 +08:00
</component>
<component name="masterDetails">
<states>
<state key="ScopeChooserConfigurable.UI">
<settings>
<splitter-proportions>
<option name="proportions">
<list>
<option value="0.2" />
</list>
</option>
</splitter-proportions>
</settings>
</state>
</states>
</component>
</project>