Xbase++ Requires Chilkat v11.7.0+

PDF to Markdown Options: Page Breaks, Images, Tables, Emphasis, Heading Levels

See more PDF Text Extraction Examples

Demonstrates the options of the "markdown" operation: pageBreaks (comment, rule or none), images (placeholder or none), tables (ruled or none), emphasis, and maxHeadingLevel. Xbase++

LOCAL nSuccess
LOCAL oPdf
LOCAL oJson
LOCAL oSbMarkdown
LOCAL oJsonOut

nSuccess := 0

//  This example requires the Chilkat API to have been previously unlocked.
//  See Global Unlock Sample for sample code.

//  The two-page PDF used in this example is available at markdown_options.pdf

oPdf := CreateObject("Chilkat.Pdf")
nSuccess := oPdf:LoadFile("qa_data/pdf/markdown_options.pdf")
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    RETURN
ENDIF

oJson := CreateObject("Chilkat.JsonObject")
oSbMarkdown := CreateObject("Chilkat.StringBuilder")
oJsonOut := CreateObject("Chilkat.JsonObject")

//  First with the default options.
nSuccess := oPdf:ExtractTextSb("markdown", oJson, oSbMarkdown, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oSbMarkdown:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "---- default options ----"
? oSbMarkdown:GetAsString()

//  Now change the options:

//  pageBreaks: "comment" (the default) separates pages with <!-- page N -->,
//  "rule" separates them with a horizontal rule (---), and "none" runs the pages together.
oJson:UpdateString("pageBreaks", "rule")

//  images: "placeholder" (the default) marks where each image sits with ![image](page-1-image-1);
//  "none" leaves images out.
oJson:UpdateString("images", "none")

//  tables: "ruled" (the default) converts tables that have drawn cell borders into Markdown tables;
//  "none" returns their cells as plain text.
oJson:UpdateString("tables", "none")

//  emphasis: false leaves out the ** and * markers for bold and italic.
oJson:UpdateBool("emphasis", 0)

//  maxHeadingLevel: inferred headings deeper than this level are folded into this level.
//  The PDF has three heading levels; with maxHeadingLevel 2 the third level becomes ##.
oJson:UpdateInt("maxHeadingLevel", 2)

oSbMarkdown:Clear()
nSuccess := oPdf:ExtractTextSb("markdown", oJson, oSbMarkdown, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oSbMarkdown:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "---- custom options ----"
? oSbMarkdown:GetAsString()

//  Sample output:

//  ---- default options ----
//  <!-- page 1 -->
//  
//  # Community Garden Handbook
//  
//  ## Plots
//  
//  ### Plot Sizes
//  
//  Plots come in two sizes. A **full plot** is twenty feet by ten feet; a **half plot** is ten feet square. New gardeners are *strongly encouraged* to start with a half plot.
//  
//  ![image](page-1-image-1)
//  
//  ### Plot Fees
//  
//  | Plot | Size | Annual Fee |
//  | --- | --- | --- |
//  | Full | 20 x 10 ft | $60 |
//  | Half | 10 x 10 ft | $35 |
//  
//  <!-- page 2 -->
//  
//  ## Watering
//  
//  ### Schedule
//  
//  The shared tanks are filled on Monday and Thursday mornings. Please water by hand from the tanks rather than running a hose from the street tap, which is metered.
//  
//  ### Dry Spells
//  
//  During a declared drought the committee may limit watering to **two days a week**. Notices are posted on the shed door and sent by e-mail.
//  
//  ---- custom options ----
//  # Community Garden Handbook
//  
//  ## Plots
//  
//  ## Plot Sizes
//  
//  Plots come in two sizes. A full plot is twenty feet by ten feet; a half plot is ten feet square. New gardeners are strongly encouraged to start with a half plot.
//  
//  ## Plot Fees
//  
//  Plot Size Annual Fee
//  
//  Full 20 x 10 ft $60
//  
//  Half 10 x 10 ft $35
//  
//  ---
//  
//  ## Watering
//  
//  ## Schedule
//  
//  The shared tanks are filled on Monday and Thursday mornings. Please water by hand from the tanks rather than running a hose from the street tap, which is metered.
//  
//  ## Dry Spells
//  
//  During a declared drought the committee may limit watering to two days a week. Notices are posted on the shed door and sent by e-mail.

oPdf:destroy()
oJson:destroy()
oSbMarkdown:destroy()
oJsonOut:destroy()