bug: Spark Connect Page Rank Errors #889

Describe the bug

When computing page rank via Spark Connect, we get an AttributeError: 'GraphFrameConnect' object has no attribute 'edges'.

To Reproduce

Steps to reproduce the behavior:

1. Spin up a Spark Connect Server
2. Run a basic pagerank off of a GraphFrame:

from graphframes import GraphFrame
from pyspark.sql import Row

vertices = spark.createDataFrame([
    Row(id="a", name="Alice"),
    Row(id="b", name="Bob"),
    Row(id="c", name-"Charlie"),
])

edges = spark.createDataFrame([
    Row(src="a", dst="b", relationship="friend"),
    Row(src="b", dst="c", relationship="friend"),
    Row(src="c", dst="a", relationship="friend"),
])

g = GraphFrame(vertices, edges)
pagerank = g.pageRank(resetProbability=0.15, maxIter=5)
top_node = pagerank.vertices.orderby("pagerank", ascending=False).first()

3. Get AttributeError

Expected behavior

pageRank function simply works and returns the results.

System [please complete the following information]:

• OS: MacOS Tahoe 26.5.2
• Python Version (if applied): 3.14
• Spark / PySpark version: PySpark 3.5.8
• GraphFrames version: graphframes-py 0.12.1

Component

• Scala Core Internal
• Scala API
• Spark Connect Plugin
• PySpark Classic
• PySpark Connect

Additional context

Looking at the source, it looks like we should be using self._edges versus self.edges in the GraphFrameConnect class.

Are you planning on creating a PR?

• I'm willing to make a pull-request