tests/testthat/_snaps/write-parquet-4.md

In nanoparquet: Read and Write 'Parquet' Files

errors

Code
  .Call(rf_nanoparquet_write, mtcars, tempfile(), dim(mtcars), 0L, list(character(),
  character()), rep(FALSE, ncol(mtcars)), options, map_schema_to_df(NULL, mtcars),
  rep(10L, ncol(mtcars)), 1L, sys.call())
Condition
  Error:
  ! Unknown Praquet encoding code: 10

force PLAIN / RLE

Code
  read_parquet_pages(tmp)[["encoding"]]
Output
  [1] "PLAIN" "PLAIN"

Code
  read_parquet_pages(tmp)[["encoding"]]
Output
  [1] "PLAIN"          "RLE_DICTIONARY" "PLAIN"          "RLE_DICTIONARY"

cutoff for dict encoding decision

Code
  read_parquet_pages(tmp)[["encoding"]]
Output
  [1] "PLAIN"

Code
  read_parquet_pages(tmp)[["encoding"]]
Output
  [1] "PLAIN"

write broken DECIMAL INT32

Code
  as.data.frame(read_parquet_schema(tmp))[, -1]
Output
      name  r_type  type type_length repetition_type converted_type logical_type
  1 schema    <NA>  <NA>          NA            <NA>           <NA>             
  2    dec integer INT32          NA        REQUIRED        DECIMAL             
    num_children scale precision field_id
  1            1    NA        NA       NA
  2           NA     2         5       NA

Code
  write_parquet(d, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Invalid Parquet file: scale is not set for DECIMAL converted type

Code
  write_parquet(d, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Invalid Parquet file: precision is not set for DECIMAL converted type

Code
  write_parquet(d, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Internal nanoparquet error, precision to high for INT32 DECIMAL

Code
  write_parquet(d2, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Internal nanoparquet error, precision to high for INT32 DECIMAL

write broken DECIMAL INT64

Code
  write_parquet(d, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Internal nanoparquet error, precision to high for INT64 DECIMAL

Code
  write_parquet(d2, tmp, schema = schema2)
Condition
  Error in `write_parquet()`:
  ! Internal nanoparquet error, precision to high for INT64 DECIMAL

write broken INT32

Code
  write_parquet(d, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

Code
  write_parquet(d2, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

write broken UINT32

Code
  write_parquet(d, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

Code
  write_parquet(d2, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

write broken INT64

Code
  write_parquet(d2, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT64 INT type: 32

INT96 errors

Code
  write_parquet(d, tmp, schema = parquet_schema("INT96"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet INT96 type.

FLOAT errors

Code
  write_parquet(d, tmp, schema = parquet_schema("FLOAT"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet FLOAT type.

DOUBLE errors

Code
  write_parquet(d, tmp, schema = parquet_schema("DOUBLE"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet DOUBLE type.

BYTE_ARRAY errors

Code
  write_parquet(d, tmp, schema = parquet_schema("BYTE_ARRAY"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet BYTE_ARRAY element when writing a list column of RAW vectors.

Code
  write_parquet(d2, tmp, schema = parquet_schema("BYTE_ARRAY"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet BYTE_ARRAY type.

FIXED_LEN_BYTE_ARRAY errors

Code
  write_parquet(d, tmp, schema = parquet_schema("UUID"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a list as a Parquet UUID (FIXED_LEN_BYTE_ARRAY) type.

Code
  write_parquet(d, tmp, schema = parquet_schema("FLOAT16"))
Condition
  Error in `write_parquet()`:
  ! Cannot write a list as a Parquet FLOAT16 type.

Code
  write_parquet(d2, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid string length: 6, expenting 3 for FIXED_LEN_TYPE_ARRAY

Code
  write_parquet(d, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet BYTE_ARRAY element when writing a list column of RAW vectors.

Code
  write_parquet(d3, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Invalid string length: 4, expenting 3 for FIXED_LEN_TYPE_ARRAY

Code
  write_parquet(d4, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Cannot write a logical vector as a Parquet FIXED_LEN_BYTE_ARRAY type.

BOOLEAN errors

Code
  write_parquet(d, tmp, schema = parquet_schema("BOOLEAN"))
Condition
  Error in `write_parquet()`:
  ! Cannot write an integer vector as a Parquet BOOLEAN type.

Code
  write_parquet(d, tmp, schema = schema)
Condition
  Error in `write_parquet()`:
  ! Cannot write an integer vector as a Parquet BOOLEAN type.

Errors when writing a dictionary

Code
  write_parquet(d, tmp, schema = parquet_schema("DOUBLE"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert an integer vector to Parquet type DOUBLE.

Code
  write_parquet(d2, tmp, schema = parquet_schema("BYTE_ARRAY"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a double vector to Parquet type BYTE_ARRAY.

Code
  write_parquet(d3, tmp, schema = parquet_schema("DOUBLE"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a factor to Parquet type DOUBLE.

Code
  write_parquet(d4, tmp, schema = schema4, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too small for INT32 DECIMAL with precision 2, scale 0: -101

Code
  write_parquet(d5, tmp, schema = schema5, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too large for INT32 DECIMAL with precision 2, scale 0: 101

Code
  write_parquet(d4, tmp, schema = schema4, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too small for INT64 DECIMAL with precision 2, scale 0: -101

Code
  write_parquet(d5, tmp, schema = schema5, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too large for INT64 DECIMAL with precision 2, scale 0: 101

Code
  write_parquet(d5, tmp, schema = parquet_schema("DOUBLE"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert an integer vector to Parquet type DOUBLE.

POSIXct dictionary

Code
  as.data.frame(read_parquet(tmp))
Output
                      d
  1 2024-08-20 12:38:00
Code
  read_parquet_metadata(tmp)[["column_chunks"]]$encodings
Output
  [[1]]
  [1] "PLAIN"          "RLE_DICTIONARY"

Code
  write_parquet(d, tmp, schema = parquet_schema("INT32"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a POSIXct vector to Parquet type INT32.

Code
  as.data.frame(read_parquet(tmp))
Output
         x
  1 1 secs
Code
  read_parquet_metadata(tmp)[["column_chunks"]]$encodings
Output
  [[1]]
  [1] "PLAIN"          "RLE_DICTIONARY"

Code
  write_parquet(d2, tmp, schema = parquet_schema("INT32"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a difftime vector to Parquet type INT32.

more dictionaries

Code
  write_parquet(d4, tmp, schema = schema4, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too small for INT32 DECIMAL with precision 2, scale 0: -101.000000

Code
  write_parquet(d5, tmp, schema = schema5, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too large for INT32 DECIMAL with precision 2, scale 0: 101.000000

Code
  write_parquet(d5, tmp, schema = schema, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

Code
  write_parquet(d5, tmp, schema = schema, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Invalid bit width for INT32: 64

Code
  write_parquet(d6, tmp, schema = parquet_schema("INT_8"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Integer value too large for INT with bit width 8: 128.000000 at column 1

Code
  write_parquet(d7, tmp, schema = parquet_schema("UINT_8"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Integer value too large for INT with bit width 8: 256.000000 at column 1.

Code
  write_parquet(d8, tmp, schema = parquet_schema("UINT_8"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Negative values are not allowed in unsigned INT column: -1.000000 at column 1.

Code
  write_parquet(d9, tmp, schema = schema4, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too small for INT64 DECIMAL with precision 2, scale 0: -101.000000

Code
  write_parquet(d10, tmp, schema = schema5, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Value too large for INT64 DECIMAL with precision 2, scale 0: 101.000000

Code
  write_parquet(d10, tmp, schema = parquet_schema("BYTE_ARRAY"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a double vector to Parquet type BYTE_ARRAY.

Even more dictionaries

Code
  write_parquet(d, tmp, schema = parquet_schema("UUID"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Invalid UUID value in column 1

Code
  write_parquet(d2, tmp, schema = sch2, encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Invalid string length in FIXED_LEN_BYTE_ARRAY column: 9, should be 3.

Code
  write_parquet(d2, tmp, schema = parquet_schema("DOUBLE"), encoding = "RLE_DICTIONARY")
Condition
  Error in `write_parquet()`:
  ! Cannot convert a character vector to Parquet type DOUBLE.

R -> Parquet mapping error

Code
  infer_parquet_schema(d)
Condition
  Error in `write_parquet()`:
  ! Cannot map a raw vector to any Parquet type

argument errors

Code
  write_parquet(mtcars, 1:10)
Condition
  Error in `path.expand()`:
  ! invalid 'path' argument
Code
  write_parquet(mtcars, letters)
Condition
  Error in `write_parquet()`:
  ! nanoparquet_write: filename must be a string

nanoparquet documentation built on April 3, 2025, 11:26 p.m.