R语言：基于Condition条件高效分类数据框日期为节假日/正常日

Hey there! Let’s fix this date classification problem efficiently in R—no more slow for loops or mismatched conditions.

First, let’s recap your rules to make sure we’re aligned:

• If a date exists in dfholidays and Condition == 1: mark as Holidays
• If a date exists in dfholidays but Condition == 0: mark as Normal T.1
• If a date isn’t in dfholidays at all: mark as Normal T.2

The core issue with your earlier ifelse attempt was likely that you weren’t linking each date to its corresponding Condition value first. Just checking if a date is in dfholidays doesn’t carry over the Condition data—so we need to join the two data frames first, then apply our rules.

Solution 1: Using dplyr (most readable & efficient for most cases) #

This uses vectorized operations (way faster than loops) and clear conditional logic with case_when:

library(dplyr)

# First, make sure your date columns are actual Date types (critical for matching!)
df_main$date <- as.Date(df_main$date)
dfholidays$date <- as.Date(dfholidays$date)

# Left join to bring in Condition values for matching dates
df_classified <- df_main %>%
  left_join(dfholidays, by = "date") %>%
  # Apply your classification rules
  mutate(
    date_type = case_when(
      Condition == 1 ~ "Holidays",
      Condition == 0 ~ "Normal T.1",
      is.na(Condition) ~ "Normal T.2"  # No match in dfholidays
    )
  ) %>%
  # Optional: Remove the Condition column if you don't need it anymore
  select(-Condition)

Solution 2: Base R (no extra packages needed) #

If you prefer sticking to base R, this works just as well:

# Ensure date columns are Date types
df_main$date <- as.Date(df_main$date)
dfholidays$date <- as.Date(dfholidays$date)

# Left join to merge Condition data
df_joined <- merge(df_main, dfholidays, by = "date", all.x = TRUE)

# Apply classification with nested ifelse
df_joined$date_type <- with(df_joined,
  ifelse(Condition == 1, "Holidays",
    ifelse(Condition == 0, "Normal T.1", "Normal T.2")
  )
)

# Optional: Clean up the Condition column
df_joined <- df_joined[, !names(df_joined) %in% "Condition"]

Solution 3: data.table (for huge datasets) #

If you’re working with 100k+ rows, data.table will be even faster than dplyr:

library(data.table)

# Convert to data.table objects
setDT(df_main)
setDT(dfholidays)

# Ensure date types
df_main[, date := as.Date(date)]
dfholidays[, date := as.Date(date)]

# Join, classify, and clean up
df_classified <- df_main[dfholidays, on = "date", Condition := i.Condition][
  , date_type := fcase(
    Condition == 1, "Holidays",
    Condition == 0, "Normal T.1",
    is.na(Condition), "Normal T.2"
  )
][, Condition := NULL]

Why this fixes your earlier issue: #

By doing a left join, every date in your main data frame gets paired with its exact Condition value from dfholidays (or NA if there’s no match). This means your conditional checks can directly reference the correct Condition for each date—no more mismatched classifications like your test row 4.

Quick Optimization Tips: #

1. Always use Date types: Never store dates as strings—this avoids matching errors (e.g., "2023-01-01" vs "2023/01/01").
2. Avoid for loops: R is built for vectorized operations. Loops are slow for large data because they process one row at a time, while joins/case_when process all rows at once.
3. Test with small data first: Run your code on a tiny subset of your data to verify the classification logic before scaling up.

内容的提问来源于stack exchange，提问作者alvaropr