Table of Contents

SEARCHFROMSOURCE Command Documentation

DESCRIPTION

The searchfromsource command reads a CSV source file and uses the first column (ignoring the header) as input items to search for in the vault.

For every row in the source CSV, PDMShell performs a vault search and returns information such as:

  • File ID
  • Full vault path
  • Parent folder ID
  • Checked-out status
  • Where Used parent file IDs

Optionally, the command can write the results to a CSV file and add/update that CSV in the vault.

SYNTAX

searchfromsource -filePath -recursive -csv

PARAMETERS

filePath

Path to the source CSV file to read input values.

  • If the path is absolute, PDMShell uses it directly.
  • If the path is relative, PDMShell combines it with the current directory.
Important

The source file must exist in the vault (so it can be downloaded locally and read).

recursive

If specified, the search will recursively search through all subfolders.

  • When omitted: search is performed in the current directory scope (non-recursive).
  • When included: recursive search is enabled.

csv

Optional output CSV file name/path.

If provided, the command will write the search results to a CSV file containing one row per input item.

Supported behaviors:

  • If the CSV file already exists in the vault, PDMShell will attempt to overwrite/update it.
  • If the CSV file does not exist, PDMShell will create it and add it to the vault.

INPUT CSV FORMAT

The command reads:

  • Comma-delimited CSV
  • First row is treated as a header and ignored
  • Only the first column is used

Example:

FileName
part1.sldprt
asm_top.sldasm
drawing1.slddrw

Notes:

Input Example 1: Plain file names (most common)

This is the simplest format where the first column contains exact file names:

FileName  
bracket.sldprt  
frame.sldasm  
cover.sldprt  
motor_mount.sldprt

How it behaves:

  • Each row is searched individually
  • PDMShell tries to locate the file in the vault (based on your current folder and -recursive flag)

Input Example 2: File names with extensions mixed (multi-document types)

You can mix parts, assemblies, drawings, PDFs, etc.

FileName  
100023.sldprt  
100023.slddrw  
Datasheet_100023.pdf  
SpecSheet.docx

How it behaves:

  • Each row is treated as a separate lookup item
  • Results are returned only if PDM finds a matching file

Input Example 3: Using wildcards (%) in the input

If your source file contains wildcard patterns, each row can be used as a search token:

Query

%.sldprt  
%.sldasm  
%.slddrw

How it behaves:

  • Each row becomes its own search query
  • Useful for generating bulk reports from common patterns

Input Example 4: Advanced Search expressions in the input

You can pass more powerful search values in the first column (same idea as -search in the search command).

Query

"Name=%.slddrw"  // all drawings 
"Name=%.sldprt;Locked=true" // All parts that are checked out

How it behaves:

  • Each row is passed directly into the search engine
  • Results depend on your Advanced Search capabilities and syntax rules
Note

For full syntax and supported fields/operators, see: advancedsearch.md

OUTPUT CSV FORMAT

When -csv is used, PDMShell generates the following columns:

ID,FileName,Path,ParentFolderID,ParentFolder,IsCheckedOut,WhereUsedIds

Column Definitions

  • ID: PDM file ID returned by search
  • FileName: The original value read from the source CSV first column
  • Path: Full vault path returned by the search result
  • ParentFolderID: Folder ID containing the file
  • ParentFolder: Folder path (derived from the result path)
  • IsCheckedOut: True/False based on PDM lock status
  • WhereUsedIds: Comma-separated list of parent file IDs (from the reference tree)

EXAMPLES

Example 1: Run search from a source CSV

searchfromsource -filePath "input.csv"

Reads input.csv and searches each row item.

searchfromsource -filePath "input.csv" -recursive

Searches for each item recursively through subfolders.

Example 3: Export results to a CSV file

searchfromsource -filePath "input.csv" -recursive -csv "results.csv"

Writes results into results.csv.
If results.csv exists in the vault, the command attempts to update it.
If it does not exist, the command creates it and adds it to the vault.

NOTES

  • This command requires the user to be logged in to a vault.
  • The input CSV must be accessible locally (PDM will download the file when needed).
  • If the vault search yields no result for a row, the command prints a warning and continues.
  • Output CSV values are CSV-escaped (commas, quotes, newlines).
  • Where Used results are generated using the reference tree lookup.

LIMITATIONS

  • Only the first search result is used (GetFirstResult()).
  • Only file results are processed (folder results are ignored).
  • The input file is interpreted as comma-delimited CSV only.
Edit this page