The image bounded by a simple unbounded Schwarz--Christoffel chain #
For integrable finite prevertices and total exponent in [-1, 1), an injective real boundary
parametrization forces the primitive's interior image to avoid the boundary. Consequently the
primitive maps the upper half-plane bijectively onto the complementary component containing the
base-point image, and its frontier is the whole boundary chain. This gives the direct mapping
theorem for simple unbounded polygons, including parallel-ended polygons.
The same holds at total exponent 1, an end of opening 2π, when the logarithmic coefficient
((∑ i, e i * a i) ^ 2 - ∑ i, e i * a i ^ 2) / 2 is negative. Some sign condition is needed
there: with exponents -1 / 2 at -1 and 3 / 2 at 1 the boundary is a simple chain of two
parallel rays joined by a segment, but the corner of opening 5π / 2 makes the image overlap
its boundary.
Inversion about an exterior point reduces separation to the planar Jordan curve theorem: the
inverted image is bounded, its frontier lies on the inverted boundary together with 0, and
an open subset of the filled hull of a Jordan curve cannot meet that curve.
References #
- L. Ahlfors, Complex Analysis, Chapter 6, Section 2.
- T. Driscoll and L. Trefethen, Schwarz--Christoffel Mapping, Chapter 2.
A simple proper Schwarz--Christoffel boundary is disjoint from the image of the open upper
half-plane, if the end at infinity has opening less than 2π, or opening 2π with negative
logarithmic coefficient. The total exponents -1 and 1 include parallel outer sides.
The image of a primitive with a simple proper boundary is the complementary component
containing its base-point image, for total exponent in [-1, 1), or total exponent 1 with
negative logarithmic coefficient.
The frontier of the primitive image is the entire simple proper boundary chain.
A Schwarz--Christoffel primitive with a simple proper boundary and total exponent in
[-1, 1), or total exponent 1 with negative logarithmic coefficient
((∑ i, e i * a i) ^ 2 - ∑ i, e i * a i ^ 2) / 2, maps the upper half-plane bijectively onto the
complementary region containing its base-point image. This allows reentrant finite corners,
parallel outer sides, and ends of opening 2π.