How to combine columns with extra strings into a concatenated string column in Polars?

Question

I am trying to add another column that will contain combination of two columns (Total & percentage) into a result column(labels_value) which look like: (Total) percentage%.

Basically to wrap bracket strings on Total column and add % string at the end of combination of these two columns.

import polars as pl

pl.Config(tbl_rows=21) # increase repr defaults

so_df = pl.from_repr("""
┌──────────────┬─────────────────────┬─────┬───────┬────────────┬────────┐
│ Flag         ┆ Category            ┆ len ┆ Total ┆ percentage ┆ value  │
│ ---          ┆ ---                 ┆ --- ┆ ---   ┆ ---        ┆ ---    │
│ str          ┆ str                 ┆ i64 ┆ i64   ┆ f64        ┆ f64    │
╞══════════════╪═════════════════════╪═════╪═══════╪════════════╪════════╡
│ Outof Range  ┆ Thyroid             ┆ 7   ┆ 21    ┆ 33.33      ┆ 33.33  │
│ Outof Range  ┆ Inflammatory Marker ┆ 2   ┆ 8     ┆ 25.0       ┆ 25.0   │
│ Outof Range  ┆ Lipid               ┆ 12  ┆ 63    ┆ 19.05      ┆ 19.05  │
│ Outof Range  ┆ LFT                 ┆ 14  ┆ 87    ┆ 16.09      ┆ 16.09  │
│ Outof Range  ┆ DLC                 ┆ 11  ┆ 126   ┆ 8.73       ┆ 8.73   │
│ Outof Range  ┆ Vitamin             ┆ 1   ┆ 14    ┆ 7.14       ┆ 7.14   │
│ Outof Range  ┆ CBC                 ┆ 2   ┆ 45    ┆ 4.44       ┆ 4.44   │
│ Outof Range  ┆ KFT                 ┆ 2   ┆ 56    ┆ 3.57       ┆ 3.57   │
│ Outof Range  ┆ Urine Examination   ┆ 1   ┆ 28    ┆ 3.57       ┆ 3.57   │
│ Within Range ┆ Thyroid             ┆ 14  ┆ 21    ┆ 66.67      ┆ -66.67 │
│ Within Range ┆ Inflammatory Marker ┆ 6   ┆ 8     ┆ 75.0       ┆ -75.0  │
│ Within Range ┆ Lipid               ┆ 51  ┆ 63    ┆ 80.95      ┆ -80.95 │
│ Within Range ┆ LFT                 ┆ 73  ┆ 87    ┆ 83.91      ┆ -83.91 │
│ Within Range ┆ DLC                 ┆ 115 ┆ 126   ┆ 91.27      ┆ -91.27 │
│ Within Range ┆ Vitamin             ┆ 13  ┆ 14    ┆ 92.86      ┆ -92.86 │
│ Within Range ┆ CBC                 ┆ 43  ┆ 45    ┆ 95.56      ┆ -95.56 │
│ Within Range ┆ KFT                 ┆ 54  ┆ 56    ┆ 96.43      ┆ -96.43 │
│ Within Range ┆ Urine Examination   ┆ 27  ┆ 28    ┆ 96.43      ┆ -96.43 │
│ Within Range ┆ Anemia              ┆ 38  ┆ 38    ┆ 100.0      ┆ -100.0 │
│ Within Range ┆ Diabetes            ┆ 22  ┆ 22    ┆ 100.0      ┆ -100.0 │
│ Within Range ┆ Electrolyte         ┆ 46  ┆ 46    ┆ 100.0      ┆ -100.0 │
└──────────────┴─────────────────────┴─────┴───────┴────────────┴────────┘
""")

I have tried below three ways and none of them worked:

(so_df
#  .with_columns(labels_value = "("+str(pl.col("Total"))+") "+str(pl.col("percentage"))+"%")
#  .with_columns(labels_value = "".join(["(",str(pl.col("Total")),") ",str(pl.col("percentage")),"%"]))
#  .with_columns(labels_value =pl.concat_str([pl.col("Total"),pl.col("percentage")])))

Desired result would be to add a new column like:

┌──────────────┐
│ labels_value │
│ ---          │
│ str          │
╞══════════════╡
│ (21) 33.33%  │
│ (8) 25%      │
│ (63) 19.05%  │
│ (87) 16.09%  │
│ (126) 8.73%  │
│ …            │

Henry Harbeck · Accepted Answer · 2025-01-02 12:40:31Z

Here are a few options:

so_df.with_columns(
    labels_value_1=pl.format("({}) {}%", "Total", "percentage"),
    labels_value_2=pl.concat_str(
        pl.lit("("), "Total", pl.lit(") "), "percentage", pl.lit("%")
    ),
    labels_value_3=(
        "("
        + pl.col("Total").cast(pl.String)
        + ") "
        + pl.col("percentage").cast(pl.String)
        + "%"
    ),
)

# Only Total, percentage and outputs shown for brevity
# ┌───────┬────────────┬────────────────┬────────────────┬────────────────┐
# │ Total ┆ percentage ┆ labels_value_1 ┆ labels_value_2 ┆ labels_value_3 │
# │ ---   ┆ ---        ┆ ---            ┆ ---            ┆ ---            │
# │ i64   ┆ f64        ┆ str            ┆ str            ┆ str            │
# ╞═══════╪════════════╪════════════════╪════════════════╪════════════════╡
# │ 21    ┆ 33.33      ┆ (21) 33.33%    ┆ (21) 33.33%    ┆ (21) 33.33%    │
# │ 8     ┆ 25.0       ┆ (8) 25.0%      ┆ (8) 25.0%      ┆ (8) 25.0%      │
# │ 63    ┆ 19.05      ┆ (63) 19.05%    ┆ (63) 19.05%    ┆ (63) 19.05%    │
# │ 87    ┆ 16.09      ┆ (87) 16.09%    ┆ (87) 16.09%    ┆ (87) 16.09%    │
# │ 126   ┆ 8.73       ┆ (126) 8.73%    ┆ (126) 8.73%    ┆ (126) 8.73%    │
# │ …     ┆ …          ┆ …              ┆ …              ┆ …              │
# │ 56    ┆ 96.43      ┆ (56) 96.43%    ┆ (56) 96.43%    ┆ (56) 96.43%    │
# │ 28    ┆ 96.43      ┆ (28) 96.43%    ┆ (28) 96.43%    ┆ (28) 96.43%    │
# │ 38    ┆ 100.0      ┆ (38) 100.0%    ┆ (38) 100.0%    ┆ (38) 100.0%    │
# │ 22    ┆ 100.0      ┆ (22) 100.0%    ┆ (22) 100.0%    ┆ (22) 100.0%    │
# │ 46    ┆ 100.0      ┆ (46) 100.0%    ┆ (46) 100.0%    ┆ (46) 100.0%    │
# └───────┴────────────┴────────────────┴────────────────┴────────────────┘

pl.format I would say is most idiomatic, and it does the casting to string for you (much like Python's f-strings). It uses {} as a placeholder. This is what I'd personally recommend out of the options presented.

For concat_str, you can add the parentheses and "%" with pl.lit (strings are parsed as column names in this function). It will cast numeric columns to string for you.

Lastly, string concatenation with the + operator does work, but you must cast the column(s) with Polars' .cast method, rather than Python's builtin str.

One thing to note is that row 2 of the output does differ slightly from your desired output. (8) 25% (expected) vs (8) 25.0% (actual). This is because the string representation of floating point columns in Polars contain at least one decimal place. If that is an issue, drop a comment and I will try to come up with a solution.

Collectives™ on Stack Overflow

How to combine columns with extra strings into a concatenated string column in Polars?

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related