MongoDB投影title字段异常：需重复指定title:1才能生效？

Hey there, let's break down why you're seeing this odd behavior with your aggregation query and how to fix it cleanly.

Why does repeating title:1 work? #

MongoDB's $project stage uses a whitelist approach by default. That means:

• Only the fields you explicitly mark with 1 (or define new computed fields for) are carried over to the next pipeline stage.
• The _id field is kept automatically unless you explicitly exclude it with _id:0.

In your query, the first $project includes title:1, so title makes it to the second stage. But your second $project only defines the normalized_rating computed field—you don't explicitly include title here. So MongoDB drops the title field (and any other unmentioned fields) before moving to the next stage. When you add title:1 again in the second $project, you're telling MongoDB to keep that field alongside the new computed one, which is why it shows up in your results.

Fixes to avoid repeating fields #

You have two solid options to clean up this query:

1. Explicitly include needed fields in the second $project #

If you want to keep the two separate $project stages, just add the fields you want to retain to the second stage:

db.movies.aggregate([
  {$match: { 
    languages: {$elemMatch : {$eq: "English"}}, 
    "imdb.rating":{$gte: 1}, 
    "imdb.votes":{$gte: 1}, 
    year: {$gte: 1990} 
  }},
  {$project: {
    scaled_votes: {$add: [
      1, 
      {$multiply: [
        9, 
        {$divide: [
          {$subtract: ["$imdb.votes",5]},
          {$subtract: [1521105,5]}
        ]}
      ]}
    ]}, 
    "imdb.rating": 1, 
    title: 1 
  }},
  {$project: {
    normalized_rating: {$avg: ["$scaled_votes", "$imdb.rating"]},
    title: 1, // Keep title
    "imdb.rating": 1 // Keep rating if needed
  }}
])

2. Merge both $project stages into one (more efficient) #

There's no need to split the projection and computed fields into two separate stages. Combining them reduces pipeline overhead and eliminates the need to repeat field definitions:

db.movies.aggregate([
  {$match: { 
    languages: {$elemMatch : {$eq: "English"}}, 
    "imdb.rating":{$gte: 1}, 
    "imdb.votes":{$gte: 1}, 
    year: {$gte: 1990} 
  }},
  {$project: {
    title: 1,
    "imdb.rating": 1,
    scaled_votes: {$add: [
      1, 
      {$multiply: [
        9, 
        {$divide: [
          {$subtract: ["$imdb.votes",5]},
          {$subtract: [1521105,5]}
        ]}
      ]}
    ]},
    normalized_rating: {$avg: ["$scaled_votes", "$imdb.rating"]}
  }}
])

This single $project stage computes both scaled_votes and normalized_rating while retaining the title and imdb.rating fields you need, all in one step.

内容的提问来源于stack exchange，提问作者Rahul Raj