☰

对文本文件的每一行运行 Python 脚本

tag-icon • tag-icon command-line bash scripts python
对文本文件的每一行运行 Python 脚本

如何在 bash 中执行以下任务？

我需要编写一个脚本，对文本文件的每一行分别运行一个 python 脚本作为输入，然后将结果保存在以读取文件的行命名的 json 文件中。

因此文本文件 10tweets.txt 如下所示：

cat 10tweets.txt

Trump on the other hand goes all in on water boarding AND some. #GOPDebate
RT @wpjenna Donald Trump promises that he will not touch the 2nd amendment -- "unless we're going to make it stronger."
Trump 23%, Rubio 19%, Kasich & Bush 14%, Christie 10%, Cruz 9% #NHPrimary
@realDonaldTrump Thank you for saying you won't use vulger language anymore. Talk about Sanders & Clinton. Take Cruz as VP. Mexican votes!!!
RT @SurfPHX Mr. Trump @realDonaldTrump tweeted 25 minutes ago. You all do realize, that our future President hardly sleeps. He's a Fighter and a Worker!
go, Bernie #DemDebate
Sanders calls out Clinton on taking Foreign Policy advice from Warmonger Henry Kissinger https://t.co/xT5J4uh4m4 via @YouTube
Cruz, Rubio, and the Moral Bankruptcy of Progressive Identity Politics https://t.co/kSQstJXtKO via @NRO
RT @scarylawyerguy "Who does Bernie Sanders listen to on foreign policy." - A question Hillary had to raise b/c the media will not. #DemDebate
Why Did U of California Fire Tenured Riverside Professor? / Ted Cruz and Higher Ed -- ... - https://t.co/zFxa4Q70wh

我希望输出文件夹中的输出像 1.json、2.json、3.json、4.json、5.json。

不确定如何在 bash 脚本中使用exec entity_sentiment.py "$@"并将其链接到文件中的每一行。脚本的运行方式如下

$ python entity_sentiment.py sentiment-entities-text "Thank you for saying you won't use vulger language anymore"
Mentions: 
Name: "vulger language"
  Begin Offset : 35
  Content : vulger language
  Magnitude : 0.699999988079071
  Sentiment : -0.699999988079071
  Type : 2
Salience: 1.0
Sentiment: magnitude: 0.699999988079071
score: -0.699999988079071

例如，脚本的输入可以假定为文件的第一行。

基本上，运行以下 bash 脚本只会分析文件的最后一行并将其保存在 1.json 中

#!/bin/bash

n=1

while read -u 3 -r line; do
  python entity_sentiment.py sentiment-entities-text "$line" > "$((n++)).json"
done 3< 10tweets.txt

下面是我在 bash IRC 频道中运行建议时发生的片段：https://pastebin.com/raw/VQpPFJYs和https://pastebin.com/raw/GQefrTX0

答案1

您可以read逐行输入文件，并使用类似以下命令将命令应用于每一行

#!/bin/bash

n=1

while read -r line; do
  python entity_sentiment.py sentiment-entities-text "$line" > "$((n++)).json"
done < input.txt

我不明白这exec entity_sentiment.py "$@"会有什么帮助。

答案2

感谢IRC bash社区

#!/bin/bash

n=1

while read -u 3 -r line; do
  echo $n "${line::30}"
  python entity_sentiment.py sentiment-entities-text "$line" > "$((n++)).json"
  ((n++))
done 3< 10tweets.txt

相关内容